DAC listening test files - update

[pma]

Test survey and test results as of August 23, 2026 ... and later

results with p <= 0.05 are typed in bold characters. Green pass is for more than one result with p<=0.05. Calculated p-values for each participant in each test added.
pdf files are attached at the bottom of this post

----------------------------------------------------------------------------------------------------------------------------------------------------------

Audibility of DACs has been discussed in another thread and I have been asked to post the test files in the new thread. So I am opening this new thread and posting some info:

Part I. Original file vs. DAC/ADC (Topping DX5 / E1DA Cosmos)

Original info was posted here. There are two 96kHz/24bit wav files in the Dactest.zip file. The zip file is to be downloaded and unzipped. I did my best to match the files in level and time alignment. The link to download the files is:

Dactest.zip

drive.google.com

Feel free to listen to or analyze the files, but most of all I will appreciate a foobar abx protocol report, to show if there was or was not an audible difference.

Update August 8, 2026
Part II. Topping DX5 vs. Topping D10s

DX5_D10s.zip

drive.google.com

DeltaWave reports below, Dactest is original vs. DX5

Update August 11, 2026
Part III. Vincent CD-S3 vs. DacMagic+

alphabeta.zip

drive.google.com

Part IV. Topping D10s vs. DacMagic+

https://pmacura.cz/2_DAC_test.zip

A/B preamp switch was used

Online test Original data vs. Marantz SA7001 output

Marantz SA7001 CD player output vs. original CD data · FunkyABX

abx.funkybits.fr

This is an online test that I prepared in March 2026. Wav files. Anyone can try it online, but the test is closed
for the reason of cheating. However, the files are available for download in this post.
---------------------------------------

So far we have spoken about tinniest differences between DACs "sound. So have about trying audibility of a poor tube preamp in the DAC/ADC loop? This is the link to my new thread and this to the new test (Loop vs. Tube preamp):

6J1_test.zip

drive.google.com

and online test:

6J1 tube preamp vs. Topping DAC/Cosmos ADC loop · FunkyABX

abx.funkybits.fr

 

[pjug]

Thanks for doing this. I will ABX this when I have a chance today.

It looks like maybe you started with a good DAC and your worst one? Maybe you don't want to say yet. Anyway that is a good idea since I feel I have some chance of passing an ABX, or if not me I bet someone here will pass it.

 

[NTTY]

Thank you! Do you want us to provide what hardware we use for listening?
I suppose I’ll test anyways with couple of DACs / HP amps.

 

[pma]

Thank you! Do you want us to provide what hardware we use for listening?
I suppose I’ll test anyways with couple of DACs / HP amps.

At the moment I prefer to keep everything undisclosed, because any disclosure leads to prejudice and preliminary judgements. In case we collect some feedback and listening test results, I will definitely disclose the test conditions.

 

[Thomas_A]

Nice to see. However, I’ve done enough blind tests in the past, and unless some unit is broken or is intentionally made poor, I know I will not pass. It needs to be either polarity flipping or very slow filtering (like >= 1 dB drop at 10k) to me if it should be audible. At the age of 59 I am not expecting it to improve from those past blind tests…

 

[pma]

At the age of 59 I am not expecting it to improve from those past blind tests…

Yeah I know. But you are young, I am 71 .

 

[pma]

I suppose I’ll test anyways with couple of DACs / HP amps.

Looking forward your findings, and if I may ask, as you are very experienced tester - in case you have some important findings, please send it to me in a PM, rather than writing it here directly, for us not to influence other readers.

 

[pjug]

I started a little bit with training mode in Foobar2000. I am not hearing anything to focus on to help me pass the ABX. When someone points me to some clues in the tracks to listen for, then my ears and concentration tend to be good enough that I can pass an ABX when I should be able to. But if I have to find the clues on my own I am bad at that.

 

[JSmith]

help me pass the ABX

Harman How to Listen

A blog for users of the listener training software application "How to Listen."

harmanhowtolisten.blogspot.com

JSmith

 

[RandomEar]

I'll start with the results: 6/16. So, in essence, you could say I'm pretty good at not identifying the differences! I agree with @pjug that until now, I didn't find any clue or specific region to listen to. Maybe that doesn't even exist for these two files. We will see!

 

[pjug]

Harman How to Listen

A blog for users of the listener training software application "How to Listen."

harmanhowtolisten.blogspot.com

JSmith

Maybe I should do that or similar but it will take time. For now I hope others will try this, and I will keep trying to hear some difference before doing the ABX myself.

 

[Matt_Holland]

Audibility of DACs has been discussed in another thread and I have been asked to post the test files in the new thread. So I am opening this new thread and posting some info:

Original info was posted here. There are two 96kHz/24bit wav files in the Dactest.zip file. The zip file is to be downloaded and unzipped. I did my best to match the files in level and time alignment. The link to download the files is:

Dactest.zip

drive.google.com

Feel free to listen to or analyze the files, but most of all I will appreciate a foobar abx protocol report, to show if there was or was not an audible difference.

As this is a new thread and to make sure everything is clear for those who haven’t read through the other thread, or don’t have time to, could you include a brief summary in the OP as to what these files are and how to compare them?

Are we comparing two identical files converted via different DACs (and back to digital again)? Or are the files two different pieces of content that we should use to test two different DACs that we possess ourselves?

 

[pjug]

As this is a new thread and to make sure everything is clear for those who haven’t read through the other thread, or don’t have time to, could you include a brief summary in the OP as to what these files are and how to compare them?

Are we comparing two identical files converted via different DACs (and back to digital again)? Or are the files two different pieces of content that we should use to test two different DACs that we possess ourselves?

@pma may want to put something in the OP, but since I asked him to do this I will comment.

I asked for recordings of two or more DACs to get an idea of what might be expected for differences in an ABX test like Josh tried to do. Josh used 96KHz tracks for the testing so that the reconstriction filters would not have much influence on the differences. So I asked for the same. The recordings of the DAC outputs are using the same ADC (I think Cosmos) and so any effects like the DC filter should be the same for all of recordings of the DACs.

 

[NTTY]

Looking forward your findings, and if I may ask, as you are very experienced tester - in case you have some important findings, please send it to me in a PM, rather than writing it here directly, for us not to influence other readers.

Sure

 

[pma]

As this is a new thread and to make sure everything is clear for those who haven’t read through the other thread, or don’t have time to, could you include a brief summary in the OP as to what these files are and how to compare them?

Are we comparing two identical files converted via different DACs (and back to digital again)? Or are the files two different pieces of content that we should use to test two different DACs that we possess ourselves?

Hi, the files are not identical, the goal is to check if two different DACs may “sound” different, in case they have “good enough” parameters. For this reason I have chosen a method, which I do not wish to disclose immediately. If someone would be able to hear a sound difference which would be supported by a valid abx protocol, then we may discuss what is the main reason/parameter. You may use either the complete music or to concentrate to a selected part, foobar abx makes it possible.

You download the foobar abx component, install it, make a choice of the 2 files that I created and start the foobar ABX. To get statistically valid positive result, you need to get 12/16 or better result. 12 right choices of 16 trials. 16 trials are requested to avoid a random result. 8 trials is definitely not enough, 10 is so-so.

foobar2000:Components/ABX Comparator (foo_abx) - Hydrogenaudio Knowledgebase

wiki.hydrogenaudio.org

foobar2000: Components Repository - ABX Comparator

foobar2000

www.foobar2000.org

foobar2000 ABX Log Signature Tool

 

[Matt_Holland]

Hi, the files are not identical, the goal is to check if two different DACs may “sound” different, in case they have “good enough” parameters. For this reason I have chosen a method, which I do not wish to disclose immediately. If someone would be able to hear a sound difference which would be supported by a valid abx protocol, then we may discuss what is the main reason/parameter. You may use either the complete music or to concentrate to a selected part, foobar abx makes it possible.

You download the foobar abx component, install it, make a choice of the 2 files that I created and start the foobar ABX. To get statistically valid positive result, you need to get 12/16 or better result. 12 right choices of 16 trials. 16 trials are requested to avoid a random result. 8 trials is definitely not enough, 10 is so-so.

foobar2000:Components/ABX Comparator (foo_abx) - Hydrogenaudio Knowledgebase

wiki.hydrogenaudio.org

foobar2000: Components Repository - ABX Comparator

foobar2000

www.foobar2000.org

foobar2000 ABX Log Signature Tool

Thank you for this. Much appreciated.

 

[NTTY]

I agree Foobar ABX is really good at ABX because you can select only a section of your choice and replay A,B or X as much as you want before deciding if you heard A or B. You can use that to train yourself and search at different clues.

That said, the conditions are extreme in that you can take as much time as you want to find a difference. And if you have to do so, it means in real life, when listening to random musical content, you would not repeat your findings.

Ideally, you should have only a limited amount of time, but that’s nearly impossible to organize online without spending much time and efforts. Foobar will show time stamp in the output file, but they can be modified manually.

@pma As opposed to the video, I would use replay gain (off by default). But you said you aligned the gain of the two tracks, right?

 

[pma]

As opposed to the video, I would use replay gain (off by default). But you said you aligned the gain of the two tracks, right

Agreed of course, do not use the replay gain (I did not watch the whole video). The files are level matched in the 0.01 dB order. And time aligned to 0 samples difference.

 

[ATTI]

What ADC did you use?

 

[SSS]

Audibility of DACs has been discussed in another thread and I have been asked to post the test files in the new thread. So I am opening this new thread and posting some info:

Original info was posted here. There are two 96kHz/24bit wav files in the Dactest.zip file. The zip file is to be downloaded and unzipped. I did my best to match the files in level and time alignment. The link to download the files is:

Dactest.zip

drive.google.com

Feel free to listen to or analyze the files, but most of all I will appreciate a foobar abx protocol report, to show if there was or was not an audible difference.

Good idea. Unfortunately I can't here a significant difference between the two files even with headphones. So the question whether good external DACs sound different is not mine anymore. Maybe in my youth I heard it but at that time no consumer DACs availlable. For me internal DACs started with the CD players. And also there the difference between good ones were very small.