Audibility of SINAD differences?

[Matt_Holland]

It is just absolutely ludicrous to suggest any type of sound over YouTube or any recorded means for evaluation, experiment or comparison. I find this idea so wildly offensive to human intelligence, that to even talk about it is giving it attention that it does not deserve. In fact, I am very disappointed that Erin Hardison over at Erin's Audio Corner started to do this crap on his YouTube reviews, absolutely did damage to his reputation from my view. I will not discuss this topic any further as I do not want to be dirtied by it.

Absolutely absurdity to the 9th degree.

If you’re referring to Erin’s pink noise comparisons of speakers’ tonal balance then I think this is one of the best things a YouTube reviewer has ever done. In no way does the audio compression undermine the validity of the comparison. The compression is an experimental constant not a variable, no?

 

[PristineSound]

If you’re referring to Erin’s pink noise comparisons of speakers’ tonal balance then I think this is one of the best things a YouTube reviewer has ever done. In no way does the audio compression undermine the validity of the comparison. The compression is an experimental constant not a variable, no?

Disagreed. Absolutely dumb like a rock. How do you know if his room acoustics didn't throw off the tonal balance with standing waves and reverberations? What mic did he used, was it calibrated religiously? What if I am listening on a pair of speakers that aren't neutral, possibly with a hump on the frequency spectrum that the speakers he was recording that had a suck out?

This is a prime example of Toole's "Circle of Confusion."

This is something that subjectivist reviewers would do and it's beneath Erin.

 

[phoenixdogfan]

SINAD is useful as a relative yardstick. But don't forget that's it's a combination of distortion (which none of us are great at detecting) and noise, which we are all much more sensitive to. I tend to ignore the SINAD number and instead look at the noise behaviour.

To me, it's a ceteris paribus situation: All other things being equal, I will always prefer the component with the lowest measured noise and distortion. Thing is, for stuff like DACS and Amps most them are highly similar: Same form factor, same power output, same approximate price, same inputs and output. Given that, why agonize about what one's hearing threshold for noise and distortion are? Just pick the component that's lowest and have done with it.

 

[Sokel]

SINAD audibility is almost personal I think (noise alone not so much)

For distortion alone Distort is our best friend:

Beta Test: DISTORT - audibility of distortions

DeltaWave beta-test was a huge success, thanks to all of you! This new beta-test is again, only for the brave and the curious :) Introducing DISTORT - free software designed to help determine the audibility thresholds of various audio distortions with your own equipment, at your own pace, with...

www.audiosciencereview.com

And now the funny stuff:
You know the "sound profiles" some of the SINAD champions have?
They look like that:

That's distortion over the top.
You know the consensus about it?
No one can hear a difference, ask around, they report it in threads here too.

So...

 

[Matt_Holland]

Disagreed. Absolutely dumb like a rock. How do you know if his room acoustics didn't throw off the tonal balance with standing waves and reverberations? What mic did he used, was it calibrated religiously? What if I am listening on a pair of speakers that aren't neutral, possibly with a hump on the frequency spectrum that the speakers he was recording that had a suck out?

This is a prime example of Toole's "Circle of Confusion."

This is something that subjectivist reviewers would do and it's beneath Erin.

Hopefully you weren’t calling me dumb like a rock?!

Anyway, Erin doesn’t measure the speakers in the room whilst playing pink noise.
He measures them using his Klippel NFS to generate a true anechoic spinorama. He then convolves the data to generate a pink noise audio track that has a tonal balance like the speaker’s on-axis response (or maybe the listening window). He then adds a high pass filter at 80Hz so that the bass performance of the speakers/headphones you listen on don’t bias the result.

The aim is give listeners a relative comparison of the convolved pink noise versus normal (albeit high-passed) pink noise. Obviously it’s not an absolute test. And of course it doesn’t simulate how a speaker interacts in the room.

I think it’s a highly pragmatic solution to the constraints in place.

The dummy head stereo recordings that Ron from New Record Day did on his channel a while back were also valuable in showing speaker differences.

Both methods I believe are best listened to over high quality headphones.

 

[PristineSound]

Hopefully you weren’t calling me dumb like a rock?!

No, absolutely not. I am calling the idea of recording something and giving the content to users to play back for evaluation dumb like a rock and I stand by that statement and going to double down.

Anyway, Erin doesn’t measure the speakers in the room whilst playing pink noise.
He measures them using his Klippel NFS to generate a true anechoic spinorama. He then convolves the data to generate a pink noise audio track that has a tonal balance like the speaker’s on-axis response (or maybe the listening window). He then adds a high pass filter at 80Hz so that the bass performance of the speakers/headphones you listen on don’t bias the result.

And so? What if I played it back on this speaker? All bets are off.

Both methods I believe are best listened to over high quality headphones.

That's even a worse idea, the best science on headphone FR is based on the Harman curve which is a completely subjective reporting by test subjects and if I recalled, on a recent Soundstage podcast, a manufacturer is already disputing the validity of the Harman curve.

 

[Matt_Holland]

No, absolutely not. I am calling the idea of recording something and giving the content to users to play back for evaluation dumb like a rock and I stand by that statement and going to double down.

And so? What if I played it back on this speaker? All bets are off.

View attachment 451207


That's even a worse idea, the best science on headphone FR is based on the Harman curve which is a completely subjective reporting by test subjects and if I recalled, on a recent Soundstage podcast, a manufacturer is already disputing the validity of the Harman curve.

The point I am making is that it doesn’t matter what you playback the test on because the comparison is relative not absolute.

Erin caveats the tests explaining this.

Who else on line besides Amir and some others in this community is trying to deliver objective data and make it accessible and fun?

I think you’re being overly critical when the pros far outweigh the cons.

 

[PristineSound]

The point I am making is that it doesn’t matter what you playback the test on because the comparison is relative not absolute.

I go back to one of my previous point, if I was playing it back on a speaker that had a suck out say at 1kHz of 5dB and the speakers being recorded had a bump of 5dB at 1kHz. Then what would you hear? The frequency would be neutral at 1kHz. So even if relative, it's utterly flawed.

Toole goes into great length descripting the "Circle of Confusion," and this concept of YouTube recording is exactly what this is.

Erin caveats the tests explaining this.

Who else on line besides Amir and some others in this community is trying to deliver objective data and make it accessible and fun?

I think you’re being overly critical when the pros far outweigh the cons.

I think Erin is just trying different things to increase viewership, but this is utterly flawed, worse, it's misleading.

But anyway, I won't be too critical of him, although I expect more from him.

 

[Matt_Holland]

I go back to one of my previous point, if I was playing it back on a speaker that had a suck out say at 1kHz of 5dB and the speakers being recorded had a bump of 5dB at 1kHz. Then what would you hear? The frequency would be neutral at 1kHz. So even if relative, it's utterly flawed.

Toole goes into great length descripting the "Circle of Confusion," and this concept of YouTube recording is exactly what this is.

You hear an error as soon as the sound clip switches from the standard pink noise. You just need to know that the standard pink noise is correct and the convolved noise has relative errors.

I get where you’re coming from, but I see value in the comparisons.

 

[PristineSound]

You hear an error as soon as the sound clip switches from the standard pink noise. You just need to know that the standard pink noise is correct and the convolved noise has relative errors.

I get where you’re coming from, but I see value in the comparisons.

While no doubt an offensive practice, but hey, at least Erin's pink noise demo is less offensive than some of other idiots on YouTube like Ted Denney and Techno Dad to name a couple.