Why SINAD is not important (article)

[Saidera]

This seems to be an intelligent criticism on SINAD which was made popular on the net by AP and ASR.

It's difficult for beginners to understand it entirely, but thankfully this thread easily patched up any back door to the old fashioned supremacy of subjective testing.

 

[escksu]

I would say SINAD are important benchmarks in audio industry. However, these benchmarks themselves are insufficient to determine the actual performance of audio equipment. Let me use computer as an analogy. Benchmarks are most important in computers and almost everything can be benchmarked objectively.

SINAD are like synthetic benchmarks in computers. You might have heard of a utility called Sisoft Sandra. It comprises of numerous benchmarking tools that could test various aspects of a computer. Eg. CPU Arithmetic test (focus on CPU performance). ITs a good indication of CPU performance. However, this alone is not an indication of how well its going to perform with real world applicaiton. There are many other factors which affects real world application performance.

It would be great if a piece of music could be played through the equipment. Then a score will be generated by a computer or some testing equipment. Haha, too bad this doesn't yet exist in audio industry.

 

[solderdude]

It would be great if a piece of music could be played through the equipment. Then a score will be generated by a computer or some testing equipment. Haha, too bad this doesn't yet exist in audio industry.

There already are: DF and PK metrics. These are at least a step in that direction. Needless to say this requires a top notch ADC that won't impart its own 'signature' on top of that from the DAC. Ultimately it is not easy to properly weigh the importance of each aspect and arrive at a number that is truly the same as average or minimum detectable for all or most demanding listeners.

 

[escksu]

There already are: DF and PK metrics. These are at least a step in that direction. Needless to say this requires a top notch ADC that won't impart its own 'signature' on top of that from the DAC. Ultimately it is not easy to properly weigh the importance of each aspect and arrive at a number that is truly the same as average or minimum detectable for all or most demanding listeners.

Oh ok, nice!! Thanks for informing me!! cheers!

 

[escksu]

I just want to add that things like SINAD are important to a certain extent. Beyond a certain threshold, it doesnt matter as much. Lets take 1% THD vs 10% THD, I am dead sure almost everyone can hear the difference, don't need blind test for this. 0.1% vs 0.01%, maybe can but definitely not really obvious. 0.01% vs 0.001%, this I doubt anyone can tell. Vast majority of the audio equipment today have very low SINAD, so its not an issue.

 

[pma]

Lets take 1% THD vs 10% THD, I am dead sure almost everyone can hear the difference, don't need blind test for this. 0.1% vs 0.01%, maybe can but definitely not really obvious. 0.01% vs 0.001%, this I doubt anyone can tell.

First, SINAD is not THD. SINAD is THD+N with reverse sign.
Re your examples:
1% vs 10% might be audible and might not be (in a DBT), depending on the recording used and audio chain used.
0.1% vs 1% will be very difficult to tell in a controlled DBT. May depend on distortion profile. The result would be rather negative.
0.01% vs. 0.1% will be inaudible in a controlled DBT.
0.001% vs 0.01% will be inaudible in a controlled DBT.
The necessary condition is that all other parameters are equal (FR, Zout) and there is no clipping and no oscillations.

 

[escksu]

First, SINAD is not THD. SINAD is THD+N with reverse sign.
Re your examples:
1% vs 10% might be audible and might not be (in a DBT), depending on the recording used and audio chain used.
0.1% vs 1% will be very difficult to tell in a controlled DBT. May depend on distortion profile. The result would be rather negative.
0.01% vs. 0.1% will be inaudible in a controlled DBT.
0.001% vs 0.01% will be inaudible in a controlled DBT.
The necessary condition is that all other parameters are equal (FR, Zout) and there is no clipping and no oscillations.

Thanks for correcting me.

 

[Dealux]

I think the article/video conveniently glosses over the fact that the source gear that's pushing 115+ SINAD doesn't have audible odd harmonic distortion either. In those cases, it's usually the 2nd harmonic that has the highest peak. Also, very rarely does a DAC or amp have only a strong 2nd harmonic and low or near zero higher harmonic components if the SINAD is particularly low. The stuff that has poor SINAD usually has lots of high order harmonic distortion as well.

In a way they seem to be making an attempt to trash cheap gear that has high SINAD because of some imagined distortion components that don't show up on a graph. These people are in effect the victims of confirmation bias. They think cheap gear must sound bad because it's cheap even though there's no real explanation as to what makes high end source gear high end as far as they can tell.

 

[KeithPhantom]

trash cheap gear that has high SINAD because of some imagined distortion components that don't show up on a graph.

We can always rely on the most absolute of the tests, the null test if that was the case.

 

[preload]

Single score to rank devices is meaningless and brings useless results. Deep analysis is always needed.

I give this post a 7.6.

 

[pinyuszka]

I just want to add that things like SINAD are important to a certain extent. Beyond a certain threshold, it doesnt matter as much. Lets take 1% THD vs 10% THD, I am dead sure almost everyone can hear the difference, don't need blind test for this. 0.1% vs 0.01%, maybe can but definitely not really obvious. 0.01% vs 0.001%, this I doubt anyone can tell. Vast majority of the audio equipment today have very low SINAD, so its not an issue.

THD - The Extended Set (125Hz)

Testing Total Harmonic Distortion below the 1000Hz standard.

www.audiocheck.net

this great test confirms what has been described

 

[tuga]

Makes sense. But if you want a single score to rank devices, what do you use?

A pink panther?

 

[tuga]

My problem with articles like this is that they don't solve the problem. It's not the matter of intention.

First is that there are many profound indications of a single SINAD number(let alone full FFT of 1khz tone in the dashboard) in different components in different systems. It's not a simple number for audibility. A very straightforward problem is to use a dac as preamp feeding the power amp.

By using logic what is important instead of SINAD. Not mentioned here. Does it mean nothing matters as long as -80dB THD+N is met or does it mean one should use their ears to decide? Neither is correct by the way.

The issue is clearly most of the time is mixed of half science and half ignorance/misunderstanding. This actually can be more harmful in some way than blatant audiofoolery. Well we can't avoid it in some way because it's natural. When we know a little bit of something we like to express it. But while we are here someone had to clear it up like exactly what happened.

But on the other hand, if someone actually reads the review word by word. I think it's clearly different from just looking at the SINAD number. Interpretation needs experience and knowledge. It's the way for headphones measurements and it's the way for electronics. SINAD is just a measurement metric, It's how you use it and interpret it matters. And of course other measurements matter too.

Is there no merit in highlighting the problem in a structured and supported manner?

Most people cannot read/interpret measurements and thus won't look at them.
Of those, subjectivity-driven audiophiles will want a listening assessment report and a rating, and objectivity-driven audiophiles will want a technical assessment conclusion and a rating. If SINAD is that rating then perhaps it is a problem.

 

[Purité Audio]

SINAD really seems to upset the subjectivists, why is that, when it is just one metric?
Keith

 

[pma]

Post in thread 'Hypex UcD180HG HxR amplifier module analysis and review'
https://audiosciencereview.com/foru...-module-analysis-and-review.27852/post-968917

My 12 cents to show why the 1kHz/5W SINAD measured into resistor load is a wrong measure of amplifier comparison.

 

[jcarrico]

Mr. Silky Smooth selling his Chinese DACs because of this headphones.com video.

 

[RHO]

Mr. Silky Smooth selling his Chinese DACs because of this headphones.com video.

What a genius! He's annoyed that they all have the same names for the output filters.
Yeah, because adding confusing names will be very helpful?
...I stopped watching after that...

 

[Slick Minkus]

SINAD really seems to upset the subjectivists, why is that, when it is just one metric?
Keith

Because products they've already spoken highly of have since received relatively poor SINAD scores here?
The single metric is then used to confront and challenge their opinions into a corner that's very difficult to fight out of, especially without an extended understanding of the other measurements that can potentially backup their views?
From a pure subjectivists perspective I imagine these challenges rarely feel noble, and can be quite alienating. Especially when you feel you're being villainised or ridiculed in the process? In that position, why wouldn't you react negatively and counter with sensationalised content? Why would you waste thousands of hours studying if gratification is just a Youtube upload away?
At the end of the day, the functions of these products are not the art - the art is what they serve to present.

(With that said, I believe this headphone.com challenge is the opposite of the above. It's discourse that serves to challenge and potentially evolve understanding for all, by 8 pages so far. Ill-informed or not - that's science.)

 

[confucius_zero]

they've already spoken highly of have since received relatively poor SINAD scores here?

Perhaps those "audio enthusiasts" prefer low dynamic range equipment since that range of the audible spectrum is more perceived?

 

[Slick Minkus]

Perhaps those "audio enthusiasts" prefer low dynamic range equipment since that range of the audible spectrum is more perceived?

Meaning higher distortion and noise in the audible spectrum? Higher distortion and noise that can be observed as warmth and described as natural sounding?