Blind test - can you tell the class D amp from original loop in ABX

[Blumlein 88]

Hmm, in this particular abx, file 2 is slightly louder. Did you volume match before abx? If you did, which app did you use to adjust volume?

I've not listened to these yet. But yes I check volume matching. I'd use Audacity, select a few sections and see what the RMS level was. Making sure I compare the same sections. You can go with normalize in some cases.

 

[pma]

For the SINAD aficionados, here is the distortion plot into the 6.8 resistor used. As a reminder, the peak in the recorded sample corresponds to some 120W/6.8 ohm sine power.

THD and THD+N vs. frequency at 20W/6.8 ohm

Re volume matching, there will always be a mismatch of 0.01dB order, depending on the level where the matching was done.

This is statistics of both files.

The mismatch here is 0.06dB in Average RMS Power. This is below audibility threshold. Absolute level matching is possible only if the transfer function is ideally straight.

 

[pma]

This was the first time I used the ABX plugin on foobar, I didn't use the replay gain option, should I have used it? Should I always use replay gain for some other tests I have in mind (mp3 vs flac and such)?

No, you should not use a replay gain. Normal listening conditions should be maintained.

 

[Pdxwayne]

No, you should not use a replay gain. Normal listening conditions should be maintained.

Interesting.....

I did one without replay gain last Saturday and I got 6/8. I thought the ~0.07db louder file 2 was given me unfair advantage.

Thus, all my abx yesterday were with replaygain option turned on. As you can see, song one was increase by ~0.07db in loudness as compared to song 2.

So, what does that say about my results?

If I do not use replay gain, you think it would make no difference?

 

[Pdxwayne]

Deltawave screenshots.

PKMetric with match gain option enabled. Left channel:

PKMetric with match gain disable:

Does that ~0.07db volume match matter? Based on the charts, I think it matters.

 

[Blumlein 88]

Deltawave screenshots.

PKMetric with match gain option enabled. Left channel:
View attachment 165690

PKMetric with match gain disable:
View attachment 165691

Does that ~0.07db volume match matter? Based on the charts, I think it matters.

While it is enough to matter on null calculations what we know of hearing is the level differences less than .1 db either don't matter or matter very little. The evidence is they don't matter.

 

[Pdxwayne]

I tried without replay gain today.

First one 6/8, then 5/8, and 5/8 for the next two. I found my ears definitely got tired after those tests. Decided to take a break for a few hours and tried again. Last one got 6/8. So, total of 22/32. About 2.5% chance of guessing.

 

[Blumlein 88]

I tried without replay gain today.

First one 6/8, then 5/8, and 5/8 for the next two. I found my ears definitely got tired after those tests. Decided to take a break for a few hours and tried again. Last one got 6/8. So, total of 22/32. About 2.5% chance of guessing.

View attachment 165785
View attachment 165786View attachment 165787
View attachment 165788

Do you think having done it first with replay gain made it easier to perceive when you didn't use Replay gain?

 

[Pdxwayne]

Do you think having done it first with replay gain made it easier to perceive when you didn't use Replay gain?

Actually, no. To my surprise, same strategy didn't work well when I did my practice run. I ended up listened differently when doing my 32 tests.

 

[Pdxwayne]

Do you think having done it first with replay gain made it easier to perceive when you didn't use Replay gain?

BTW, since you and PMA insisted that ~0.07db difference does not matter, I might as well add all my results together.

53/82. 0.5% chance of guessing.

 

[maty]

JRiver MC v28.0.81 + all my win optimizations.

* DSP: Parametric Equalizer + Convolution -> easy test.

* DSP: PKHarmonic (H2 -120dB, H4 -150dB) + Parametric Equalizer + Convolution -> I am unable to differentiate!

BTW, the best sound is with PKHarmonic plugin

Announcing beta-test of PKHarmonic VST plugin

As discussed in a separate thread, this is a beta-test of a VST (Windows) plugin to allow you to dial-in an arbitrary amount of harmonic distortion while playing music with your favorite music player. Now you can enjoy that second harmonic without investing many thousands of $$$ into a tube...

www.audiosciencereview.com

 

[abdo123]

53/82. 0.5% chance of guessing.

??????

 

[Pdxwayne]

??????

53 correct out of 82 tries.

I used https://stattrek.com/online-calculator/binomial.aspx

I used the last number calculated:

 

[abdo123]

53 correct out of 82 tries.

I used https://stattrek.com/online-calculator/binomial.aspx

I used the last number calculated:

View attachment 165902

What you’re doing is called P-stacking and it’s generally very much frowned upon.

 

[Pdxwayne]

What you’re doing is called P-stacking and it’s generally very much frowned upon.

Same songs and same setup. What is the problem?

You have to account for ear fatigue too. What is the issue of taking breaks and do testing at different times?

 

[Blumlein 88]

You can't combine those with and without Replay gain.

 

[Pdxwayne]

You can't combine those with and without Replay gain.

Fine, which one do you prefer me to use? With or without?

 

[pma]

Inconsistency of results is a good indicator of randomness of results. Still the statistics game, no reliability. To me, no proof.

Yes and no replay gain to be used. Under a microscope we can always see something.

 

[tomtoo]

Maybe a stupid question, but if a difference is so tiny, that even people that concentrate like hell, and maybe have even better hearing than i, its hard for them to make it out, and every mix i hear is very different, why i should care?

 

[Pdxwayne]

Inconsistency of results is a good indicator of randomness of results. Still the statistics game, no reliability. To me, no proof.

Or it can mean at the edge of audibility......Does not matter too much.
; )

Anyway, with your attitude, I am done. Not going to give feedback of what I heard.
: P