• 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!

Possible to rank equipment based on what we actually hear and perceive?

gags11

Senior Member
Forum Donor
Joined
Feb 8, 2019
Messages
356
Likes
528
I have learned so much from all of you. And my science heart always loved graphs instead of words, so I appreciate the SINAD.

However, I read a few interesting threads in the forum. One arguing that harmonic distortion may not matter as much as SINAD may portray. I also have seen Amir criticizing others by bringing things that we do not perceive, like phase shift.

Knowing what we know about human hearing and perception, is it possible to come up with a ranking of things we can hear? For example, dynamic range Amir presents is 20hz to 20khz. Is noise at 20hz really important? Not understanding about it too much, it seems someone thought about this and came up with A-weighting. I don’t care if there is noise at 30khz or phase shift or harmonic distortion that is masked. I care about stuff that really makes a difference.

Residual noise was one thing that opened my eyes. To me this seems important as equipment with noise can be heard and is bad, while say harmonic distortion at -95dB may be masked. Case in example was my own miniDSP 88A that Amir reviewed. It was useless in my active setup, as one could hear hiss from a meter away.

Sorry for venting, but being a science forum, could we define threshold of audibility of THD and SNR (measured at various outputs), all within what humans perceive. We don’t need ghz snr or stuff with no phase shift, as you all have argued those don’t matter.
Cheers!

PS: We measure and define preferred headphone frequency curve based on how human ear perceives sound with headphones on the ear. We do not use Hartman curve, which we came up with based on human hearing perception and preferences. We do not rank speakers based on purely flat response to 30 or 90 kHz
 
Last edited:
OP
G

gags11

Senior Member
Forum Donor
Joined
Feb 8, 2019
Messages
356
Likes
528
A-weighted S/N and a distortion threshold should be all a person really needs.

maybe max Wattage before clipping too.

I really appreciate this and would love some of the expert engineers chiming in.
 
OP
G

gags11

Senior Member
Forum Donor
Joined
Feb 8, 2019
Messages
356
Likes
528
May be we could have a THD ranking and SNR at 3 different levels: residual, common average listening level, and full dynamic range. It would be so helpful. It would cater to everyone, as some expensive equipment reviewed here have great SNRs, but crappie THD. And I’m not talking about dedicated DACs
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
May be we could have a THD ranking and SNR at 3 different levels: residual, common average listening level, and full dynamic range. It would be so helpful. It would cater to everyone, as some expensive equipment reviewed here have great SNRs, but crappie THD. And I’m not talking about dedicated DACs

noise doesn't care about the output level, it's always there. whether the volume knob is down or up.

btw your speaker hiss, does it disappear when the MiniDSP is turned off but the speakers are on?
 
OP
G

gags11

Senior Member
Forum Donor
Joined
Feb 8, 2019
Messages
356
Likes
528
noise doesn't care about the output level, it's always there. whether the volume knob is down or up.

btw your speaker hiss, does it disappear when the MiniDSP is turned off but the speakers are on?

I agree, noise is always there, that’s why I would like to know residual background noise and dynamic range, as those 2 numbers may not be the same as I understood. May be I’m wrong though.

The hiss disappears when miniDSP is off or if I connect my other active crossovers, one of which is a cheap Behringer.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,277
Location
Gold Coast, Queensland, Australia
noise doesn't care about the output level, it's always there. whether the volume knob is down or up.

Noise is not a constant. It gets amplified by each gain stage in the system and if affected by the volume position in preamplifiers, power amplifiers and source components with variable outputs.

Most amplifiers with conventional volume pots will have a sweet spot for noise. Some are noisier at low levels than high and vice versa.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
I agree, noise is always there, that’s why I would like to know residual background noise and dynamic range, as those 2 numbers may not be the same as I understood. May be I’m wrong though.

The hiss disappears when miniDSP is off or if I connect my other active crossovers, one of which is a cheap Behringer.

residual noise and dynamic range is the same, kind of, but one value is in Volts or Vrms and the other is in dB. you can calculate dynamic range by calculating the difference between max output Vrms and residual noise Vrms.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
The problem with A weighting is that this weighting is fixed where in reality it changes with SPL (equal loudness contours)
The problem with S/N ratio is that it becomes only really important when the gain and efficiency of the transducers are fixed and high.
When one has inefficient speakers/headphones the S/N ratio that is required differs from sensitive ones not to mention the age of the listener(s).
Of course one can assume worst case scenarios and go with highest sensitivity but that might disqualify gear that may be well suited.
Then there are issues with acoustics, seal of headphones and other things that may be of importance.

If there were a single number that says it all under all circumstances it would have been invented and widely used.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
Noise is not a constant. It gets amplified by each gain stage in the system and if affected by the volume position in preamplifiers, power amplifiers and source components with variable outputs.

Most amplifiers with conventional volume pots will have a sweet spot for noise. Some are noisier at low levels than high and vice versa.

yes but when you review one component at a time there will not be any further gain stages. as a result it's pointless for a review to report residual noise over 'different output levels', it will be just the same.
 

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
My minidsp 2x4 HD also added hiss to my speakers (Paradigm Persona B). Analog in from DAC to minidsp. Minidsp to amp.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,277
Location
Gold Coast, Queensland, Australia
yes but when you review one component at a time there will not be further gain stages. as a result it's pointless for a review to report residual noise over 'different output levels', it will be just the same.

Sure, but you may have missed what I was saying. A preamplifier for instance does not have a fixed residual noise level. It depends on volume position. What volume position do you pick? Flat out- no it will never be used there. Minimum- no, that is not a use case either. So where? You pick a worst case noise level, measure it and note the position.

Some preamps get quieter the further you turn the volume, others, it is the opposite.
 
OP
G

gags11

Senior Member
Forum Donor
Joined
Feb 8, 2019
Messages
356
Likes
528
The problem with A weighting is that this weighting is fixed where in reality it changes with SPL (equal loudness contours)
The problem with S/N ratio is that it becomes only really important when the gain and efficiency of the transducers are fixed and high.
When one has inefficient speakers/headphones the S/N ratio that is required differs from sensitive ones not to mention the age of the listener(s).
Of course one can assume worst case scenarios and go with highest sensitivity but that might disqualify gear that may be well suited.
Then there are issues with acoustics, seal of headphones and other things that may be of importance.

If there were a single number that says it all under all circumstances it would have been invented and widely used.

I agree with everything you pointed out. Still, if we forget about A-weighting for now. Would it be useful having 1khz THD and SNR separately? We see this with DACs, would be interesting with amplifiers.

I agree that A-weighting is not perfect and things change with SPL. Also true SINAD is not defined only at 1khz, but we still do it. Why not come up with a reasonable weighting at 2 different SPLs and show the data? I’m sure it would be a very useful project for the world of music lovers.

Edit: that’s why we could have 3 or 4 numbers for rankings. Let people decide what is important for them.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
Sure, but you may have missed what I was saying. A preamplifier for instance does not have a fixed residual noise level. It depends on volume position. What volume position do you pick? Flat out- no it will never be used there. Minimum- no, that is not a use case either. So where? You pick a worst case noise level, measure it and note the position.

Some preamps get quieter the further you turn the volume, others, it is the opposite.

most pre-amps are also capable of outrageous voltage outputs (relative to DACs) so it's an interesting conversation to be had. but to be honest a good low noise implementation would not deviate so much to the point of going from silent to noisy just by moving the volume knob, unless it's amplifying noise upstream in the chain.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
"Possible to rank equipment based on what we actually hear and perceive?"

I suspect the answer to be "no".
 

escksu

Addicted to Fun and Learning
Joined
Jul 16, 2020
Messages
965
Likes
397
I don't think it is possible. It is because what you hear and perceived is entirely subjective. Hence, it would be down to measurements instead. However, several factors affects ranking.

1. Up to a certain point, certain measurements becomes rather meaningless. Eg. THD+N, 0.01% vs 0.001%, I can safely say no one can hear any difference between the 2. So, comparing amps with extremely low THD+N becomes impossible. The same goes for other factors like SNR etc...

2. Granted that all amps sound the same as long as they are performing within their parameters and volume is match, it is impossible to rank them.

3. Power is not really a fair point because you would have to combine it with cost and needs. You cannot simply say a 200W amp is better than a 100W amp. No doubt 200W has more power but it also cost more. Then, if the person does not need the power, its a waste.

4. If clipping happens, then I would point it out as a amp/speaking matching issue rather than inferior amp.

If you still want to go ahead any rank them, then I would say its only possible to rank them as a group rather than individually. Eg. Group A,best measurements, highest power, cost not an issue. Group B, slightly worst measurements but totally inaudible difference, less power, cheaper.. etc....
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Edit: that’s why we could have 3 or 4 numbers for rankings. Let people decide what is important for them.

Rtings works kind of like this.
The problem in practice is that the efficiency of the speaker is of importance but also room size, listening distance etc.
One would have to rate a DAC, pre-amp, power amp (load conditions ?), speaker (all aspects), circumstances it is used in.
Then one could buy all kinds of things that score high but still end up with a system that is plagued by EMC issues for instance coming from a PC.
When connecting a complete AV system sometimes unexpected noise or other noises are present that would not have been shown to be possible in separate tests.

The idea is nice but if it could be done it would already have existed.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
The conservative approach is to make noise and all distortion so low we can never hear it. It becomes a non-issue. That can be done at affordable prices for nearly all of our gear. Then you get to transducers, and that we cannot keep distortion to levels that keep it a non-issue. I sometimes think distortion of speakers/phones is all we need to worry about.
 

Chrispy

Master Contributor
Forum Donor
Joined
Feb 7, 2020
Messages
7,751
Likes
5,910
Location
PNW
I suppose you could draw up your own list, beyond that don't see how helpful it would be....
 
Top Bottom