DAC listening test files - update

[pkane]

Now the question arises if we can do this non-linear correction if our aim is to make an ABX of the DAC output vs. original or DAC1 vs. DAC2. Is such manipulation without real world audible effect? Or is it made with the only goal - to get the best possible Deltawave report, regardless possible effect to audible differences?

It's hard to say in general. Filters can be poorly constructed and some of the changes can be audible. Correcting for the filter effects corrects for frequency differences as well as phase differences due to the filter.

 

[amirm]

Example better than hundreds of explanations: Almost perfect result, hoorah, or rather not? The answer is no.

foo_abx 2.1 report
foobar2000 v2.0
2026-08-08 10:19:45

File A: file_1.wav
SHA1: 21b718084cbefae84cc1723ec49976b3c9ffa9ee
File B: file_2.wav
SHA1: 53949d842b273ee3b08edcc880344359b6e6ad50

Output:
ASIO : Focusrite USB ASIO
Crossfading: NO

10:19:45 : Test started.
10:20:26 : 01/01
10:20:40 : 02/02
10:21:19 : 03/03
10:21:33 : 04/04
10:21:46 : 05/05
10:22:00 : 06/06
10:22:14 : 06/07
10:22:29 : 07/08
10:22:44 : 08/09
10:22:58 : 09/10
10:22:58 : Test finished.

 ----------
Total: 9/10
p-value: 0.0107 (1.07%)

 -- signature --
e25daaa470c6d4f2dbd66313982523f7c7f0a3bd

And so what?? I got 9/10, p = 0.0107, by listening. But I am hiding that I had about three unsuccessful test runs before. And they all have to be taken into account. This is a random result based on not enough trials during the test.

Unless you tell us that you randomly clicked and got those results, I don't buy your argument.

I have run huge number of these tests. No way is it easy to get 8 out of 10 like you are claiming. Those results are significant enough to pay attention to. To wit, compare it to what others are posting when they can't pass the test.

You may very well have found a difference audibly and it is through that which you got these results.

 

[pma]

Unless you tell us that you randomly clicked and got those results, I don't buy your argument.

It was done by normal concentrated listening, the 9/10 individual result. My opinion is based on one of my old tests with foobar abx when I got 12/16 just clicking always to “A” button. It is randomness and probability theory, which cannot exclude any random result though statistically extremely improbable. That’s why I request repeated positive test result, not only one. So we are free to buy or not to buy each other arguments.

 

[pma]

It's hard to say in general. Filters can be poorly constructed and some of the changes can be audible. Correcting for the filter effects corrects for frequency differences as well as phase differences due to the filter.

Yes it does but still my question is if such modified file is valid to be used for evaluation and/or listening test.

 

[pma]

@Josh83 : Josh, we cannot debate here, there and everywhere. There is too much noise, misqoutes and partial quotes. I am not registered at AS forum, so the only way is to debate Here to prevent further misunderstandings. Also, I am not willing to participate in any kind of personal attacks.

 

[voodooless]

And so what?? I got 9/10, p = 0.0107, by listening. But I am hiding that I had about three unsuccessful test runs before. And they all have to be taken into account.

Yes, up to a point. With multiple runs, you’d have to take into account the fact that there might be a learning effect, and that you get better by doing more runs. In that case, the latest results should be weighted differently from the oldest ones.

In case of only 3 runs though, there really isn’t enough data to conclude such a “training effect “

 

[pma]

Correct. Most probably correct. The truth is that I have chosen a specific sample part in the last successful test based on previous failures. Still, I am not self-convinced I heard a difference reliably.

 

[voodooless]

Still, I am not self-convinced I heard a difference reliably.

Do more runs

 

[pma]

edit: also, since using the Level EQ improves the matching so much, can we conclude this ripple shows the main difference in the recordings? If

Yes. The main difference is in the passband ripple and it starts deep in the audio band at several kHz, as you have shown properly. This is how the Topping D10s filter behaves.

 

[pma]

Do more runs

I know, to your surprise . You can try as well to see how “easy” It is.

 

[voodooless]

You can try as well to see how “easy” It is.

I will at some point next week. I’m sure I won’t do better than random, but it’s always good to validate .

 

[RandomEar]

Yes. The main difference is in the passband ripple and it starts deep in the audio band at several kHz, as you have shown properly. This is how the Topping D10s filter behaves.

The amplitude seems to surpass 0.1 dB only north of 30 kHz. I would be very surprised if this has any audible effect. We will see how this whole comparator thing works out. Maybe it will be useful to test at some point.

 

[pma]

@Josh83 : I tried, at least. As you are continuing to copy and paste to AS forum, misquoting, making noise with incomplete information, I am done with you and ignoring since now.

 

[pma]

The amplitude seems to surpass 0.1 dB only north of 30 kHz. I would be very surprised if this has any audible effect. We will see how this whole comparator thing works out. Maybe it will be useful to test at some point.

You mean the comparator that is under development by @mcdn ? Or the one that Amir will maybe testing? Maybe?

 

[voodooless]

So what’s going on over at AS forum?

 

[pma]

The amplitude seems to surpass 0.1 dB only north of 30 kHz. I would be very surprised if this has any audible effect. We will see how this whole comparator thing works out. Maybe it will be useful to test at some point.

I see your reasoning.

 

[pma]

So what’s going on over at AS forum?

I think you may feel free to visit and check for yourself. I am the last one to use the copy and paste method. I hate it, personally.

 

[voodooless]

I think you may feel free to visit and check for yourself. I am the last one to use the copy and paste method. I hate it, personally.

I was more thinking about a short summary. But no worries, I’ll ask my AI fiend to do that

 

[pma]

I was more thinking about a short summary.

My summary would as well lead to partial quoting or misquoting, which I want to avoid.

 

[voodooless]

My summary would as well lead to partial quoting or misquoting, which I want to avoid.

Well, here is what the AI came up with:

pma’s test is everything Josh’s isn’t

Otherwise, let’s just say it wasn’t very impressed