DAC listening test files - update

[pma]

But at this point, this feels like desperately trying to demonstrate the difference between files. I had to condition the test variables so much to chase this result that I still seriously doubt it is useful

That is the other question. We wanted to know if the DAC/ADC record may be taken as reliable enough for comparing different DACs and sharing the results.

 

[pedrojoaquimborges]

Just out of curiosity: Could you name the time stamp?

Of course, should've done it from the get go. 0:42 to 0:45 - basically during that lead guitar phrase. Also, I could not stress enough how insignificant this difference is.

That is the other question. We wanted to know if the DAC/ADC record may be taken as reliable enough for comparing different DACs and sharing the results.

It is my opinion that, at this level of converting fidelity, it's reliable enough for telling apart lower quality converters, like the ones in older consumer audio interfaces. Don't have anywhere enough data to tell whether the same is true for high quality converters. That's why I'd love to get my hands on those files Josh was comparing.

 

[pma]

@pedrojoaquimborges : would you also like to try the second test,

DX5_D10s.zip

drive.google.com

Two DACs, and probably much more difficult to tell the difference.

 

[pedrojoaquimborges]

@pedrojoaquimborges : would you also like to try the second test,

DX5_D10s.zip

drive.google.com

Two DACs, and probably much more difficult to tell the difference.

Much more difficult. I had fully convinced myself I could easily hear a difference in tone in the snare drum and hi hat, only to get a 5/16 result. Btw it's harder to use the null test to "cheat" on this one, by finding excerpts with more difference between signals, since the clock drift is very noticeable. When listening to the null test you can clearly hear the drift as if it were a very long periodic wave modulating the amplitude.

 

[RandomEar]

My first try at the DX5 vs D10s comparison from yesterday went the same: 7/16. I tried focusing on the region with the biggest delta around 9.5 s for this run. I guess you could say I cheated and still lost the match

 

[pedrojoaquimborges]

My first try at the DX5 vs D10s comparison from yesterday went the same: 7/16. I tried focusing on the region with the biggest delta around 9.5 s for this run. I guess you could say I cheated and still lost the match

In this case (i might be wrong), I don't think you could use the delta to your advantage, since clock drift seems to be mostly responsible for the signal differences. IMO this particular test is indistinguishable for the vast majority of the population, which I guess makes it very compelling to acquire the cheaper Topping converter. EDIT: Just noticed that the DX5 features a headphone amp, so their price isn't directly comparable.

However, since both the converter units and DAC chips are made by the same manufacturers, they might share tech and algorithms, giving them similar conversion performance. It would be great to replicate this test with another unit.

 

[pma]

There is not much clock drift in those DX5 x D10s files. I do not think it might be a hint to get concentrated on.

 

[pma]

However, since both the converter units and DAC chips are made by the same manufacturers, they might share tech and algorithms, giving them similar conversion performance. It would be great to replicate this test with another unit.

Those DACs under test differ mostly in filter transfer function, resulting in delta passband ripple about 0.05dB, as shown here. Apart of that, almost no difference.

 

[NTTY]

Finally got home and found time to play with your files @pma and 3 DACs, two headphones.

I unzipped the files from DacTest.zip.
It did not try the other files (tomorrow, maybe).

Here we go:

DAC & HPA SMSL DO400, Low Gain, Beyer DT770 Pro, volume -18dB. Driver in Exclusive mode​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 17:08:04

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

Output:
Default : Speakers (SMSL USB DAC) [exclusive], 32-bit
Crossfading: NO

17:08:04 : Test started.
17:08:46 : Test restarted.
17:08:46 : 00/01
17:09:03 : Test restarted.
17:09:03 : 01/02
17:09:19 : Test restarted.
17:09:19 : 02/03
17:09:30 : Test restarted.
17:09:30 : 02/04
17:09:44 : Test restarted.
17:09:44 : 03/05
17:09:56 : Test restarted.
17:09:56 : 03/06
17:10:08 : Test restarted.
17:10:08 : 04/07
17:10:24 : Test restarted.
17:10:24 : 04/08
17:10:35 : Test restarted.
17:10:35 : 04/09
17:10:45 : Test restarted.
17:10:45 : 05/10
17:11:00 : Test restarted.
17:11:00 : 05/11
17:11:08 : Test restarted.
17:11:08 : 06/12
17:11:17 : Test restarted.
17:11:17 : 07/13
17:11:28 : Test restarted.
17:11:28 : 08/14
17:11:43 : Test restarted.
17:11:43 : 09/15
17:11:54 : Test restarted.
17:11:54 : 10/16
17:11:54 : Test finished.

----------
Total: 10/16
p-value: 0.2272 (22.72%)

-- signature --
34ca47856c30f97c2a2f21627db7507c4ae661bc

>> FAIL

Topping D50III (4Vrms – Vol=0dB) + A50III (vol at 1 o’clock) – Low Gain, Beyer DT770 Pro, exclusive mode, Filter = mode1.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 17:24:51

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

17:24:51 : Test started.
17:27:05 : Test restarted.
17:27:05 : 00/01
17:27:14 : Test restarted.
17:27:14 : 00/02
17:27:25 : Test restarted.
17:27:25 : 00/03
17:27:43 : Test restarted.
17:27:43 : 00/04
17:27:52 : Test restarted.
17:27:52 : 01/05
17:28:00 : Test restarted.
17:28:00 : 01/06
17:28:12 : Test restarted.
17:28:12 : 02/07
17:28:24 : Test restarted.
17:28:24 : 02/08
17:28:31 : Test restarted.
17:28:31 : 02/09
17:28:44 : Test restarted.
17:28:44 : 03/10
17:28:55 : Test restarted.
17:28:55 : 03/11
17:29:10 : Test restarted.
17:29:10 : 04/12
17:29:59 : Test restarted.
17:29:59 : 04/13
17:31:50 : Test restarted.
17:31:50 : 04/14
17:32:00 : Test restarted.
17:32:00 : 05/15
17:32:31 : Test restarted.
17:32:31 : 05/16
17:32:31 : Test finished.

----------
Total: 5/16
p-value: 0.9616 (96.16%)

-- signature --
aa011171214da04dd145e5209be54a1763a88151

>> FAIL

Topping D50III (4Vrms – Vol=0dB) + A50III (vol at 1 o’clock) TRS-TRS – Low Gain, Beyer DT770 Pro, exclusive mode, Filter = mode3.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 17:37:13

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

17:37:13 : Test started.
17:38:04 : Test restarted.
17:38:04 : 00/01
17:38:41 : Test restarted.
17:38:41 : 01/02
17:38:49 : Test restarted.
17:38:49 : 01/03
17:38:57 : Test restarted.
17:38:57 : 01/04
17:39:06 : Test restarted.
17:39:06 : 01/05
17:39:18 : Test restarted.
17:39:18 : 02/06
17:39:27 : Test restarted.
17:39:27 : 02/07
17:39:37 : Test restarted.
17:39:37 : 02/08
17:39:45 : Test restarted.
17:39:45 : 03/09
17:39:57 : Test restarted.
17:39:57 : 03/10
17:40:14 : Test restarted.
17:40:14 : 04/11
17:40:32 : Test restarted.
17:40:32 : 04/12
17:40:46 : Test restarted.
17:40:46 : 04/13
17:40:58 : Test restarted.
17:40:58 : 05/14
17:41:14 : Test restarted.
17:41:14 : 06/15
17:41:24 : Test restarted.
17:41:24 : 06/16
17:41:24 : Test finished.

----------
Total: 6/16
p-value: 0.8949 (89.49%)

-- signature --
4c508e68b8f1935142a6a79cd1b5a0db0017fceb


>> FAIL

Topping D50III (4Vrms – Vol=0dB) + A50III (vol at 12 o’clock) TRS-TRS – Low Gain, AudioTechnica ATH-M50X, exclusive mode, Filter = mode3.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 18:10:49

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

18:10:49 : Test started.
18:12:10 : Test restarted.
18:12:10 : 00/01
18:12:19 : Test restarted.
18:12:19 : 01/02
18:12:29 : Test restarted.
18:12:29 : 01/03
18:12:47 : Test restarted.
18:12:47 : 01/04
18:12:55 : Test restarted.
18:12:55 : 01/05
18:13:08 : Test restarted.
18:13:08 : 01/06
18:13:21 : Test restarted.
18:13:21 : 02/07
18:13:31 : Test restarted.
18:13:31 : 03/08
18:13:41 : Test restarted.
18:13:41 : 03/09
18:13:49 : Test restarted.
18:13:49 : 04/10
18:13:58 : Test restarted.
18:13:58 : 04/11
18:14:10 : Test restarted.
18:14:10 : 05/12
18:14:22 : Test restarted.
18:14:22 : 06/13
18:14:34 : Test restarted.
18:14:34 : 07/14
18:14:48 : Test restarted.
18:14:48 : 08/15
18:15:11 : Test restarted.
18:15:11 : 08/16
18:15:11 : Test finished.

----------
Total: 8/16
p-value: 0.5982 (59.82%)

-- signature --
c2eaf27cff77efcde0d0b9c4ad307e04e5dd04f3

>> FAIL

Topping D50III (4Vrms – Vol=-3dB) + A50III (vol at 12 o’clock) TRS-TRS – Low Gain, AudioTechnica ATH-M50X, exclusive mode, Filter = mode3.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 18:19:08

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

18:19:08 : Test started.
18:19:51 : Test restarted.
18:19:51 : 01/01
18:20:03 : Test restarted.
18:20:03 : 02/02
18:20:12 : Test restarted.
18:20:12 : 03/03
18:21:15 : Test restarted.
18:21:15 : 03/04
18:21:25 : Test restarted.
18:21:25 : 04/05
18:21:35 : Test restarted.
18:21:35 : 04/06
18:21:45 : Test restarted.
18:21:45 : 05/07
18:21:55 : Test restarted.
18:21:55 : 06/08
18:22:05 : Test restarted.
18:22:05 : 07/09
18:22:13 : Test restarted.
18:22:13 : 08/10
18:22:21 : Test restarted.
18:22:21 : 08/11
18:22:28 : Test restarted.
18:22:28 : 09/12
18:22:36 : Test restarted.
18:22:36 : 10/13
18:22:45 : Test restarted.
18:22:45 : 11/14
18:22:57 : Test restarted.
18:22:57 : 12/15
18:23:05 : Test restarted.
18:23:05 : 13/16
18:23:05 : Test finished.

----------
Total: 13/16
p-value: 0.0106 (1.06%)

-- signature --
d36bb27bb9574dba2294dd7737f3559275980c7f


>> SUCCESS (A massive one, actually)

Success, really? Only because the DACs is set to 3dB lower output? Preventing intersample overs, maybe? Nah, pure luck, I randomly clicked, honestly. I should have taken a lottery ticket.

Let's try again the exact same:

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 18:25:59

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

18:25:59 : Test started.
18:26:08 : Test restarted.
18:26:08 : 00/01
18:26:17 : Test restarted.
18:26:17 : 01/02
18:26:26 : Test restarted.
18:26:26 : 02/03
18:26:33 : Test restarted.
18:26:33 : 02/04
18:26:45 : Test restarted.
18:26:45 : 03/05
18:27:01 : Test restarted.
18:27:01 : 03/06
18:27:07 : Test restarted.
18:27:07 : 03/07
18:27:17 : Test restarted.
18:27:17 : 03/08
18:27:32 : Test restarted.
18:27:32 : 03/09
18:27:42 : Test restarted.
18:27:42 : 04/10
18:27:51 : Test restarted.
18:27:51 : 05/11
18:28:04 : Test restarted.
18:28:04 : 06/12
18:28:16 : Test restarted.
18:28:16 : 07/13
18:28:25 : Test restarted.
18:28:25 : 08/14
18:28:32 : Test restarted.
18:28:32 : 09/15
18:28:43 : Test restarted.
18:28:43 : 09/16
18:28:43 : Test finished.

----------
Total: 9/16
p-value: 0.4018 (40.18%)

-- signature --
0cf4867e22749e0f776acd9c9fc6ceadf6e3cf43

>>FAIL (I should have taken that lottery ticket, stupid me)

---

The wife experiment

I like to get my wife involved, she's patient enough with me.

After two replies she said "I'm bored, I can't tell a difference. Get me the same with Queen or Tears For Fears, and then we talk again".

 

[pma]

Thank you, @NTTY . And give my best wishes to your wife, for her patience.

 

[Salt]

Finally got home and found time to play with your files @pma and 3 DACs, two headphones.

Here we go:

View attachment 550434

DAC & HPA SMSL DO400, Low Gain, Beyer DT770 Pro, volume -18dB. Driver in Exclusive mode​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 17:08:04

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

Output:
Default : Speakers (SMSL USB DAC) [exclusive], 32-bit
Crossfading: NO

17:08:04 : Test started.
17:08:46 : Test restarted.
17:08:46 : 00/01
17:09:03 : Test restarted.
17:09:03 : 01/02
17:09:19 : Test restarted.
17:09:19 : 02/03
17:09:30 : Test restarted.
17:09:30 : 02/04
17:09:44 : Test restarted.
17:09:44 : 03/05
17:09:56 : Test restarted.
17:09:56 : 03/06
17:10:08 : Test restarted.
17:10:08 : 04/07
17:10:24 : Test restarted.
17:10:24 : 04/08
17:10:35 : Test restarted.
17:10:35 : 04/09
17:10:45 : Test restarted.
17:10:45 : 05/10
17:11:00 : Test restarted.
17:11:00 : 05/11
17:11:08 : Test restarted.
17:11:08 : 06/12
17:11:17 : Test restarted.
17:11:17 : 07/13
17:11:28 : Test restarted.
17:11:28 : 08/14
17:11:43 : Test restarted.
17:11:43 : 09/15
17:11:54 : Test restarted.
17:11:54 : 10/16
17:11:54 : Test finished.

----------
Total: 10/16
p-value: 0.2272 (22.72%)

-- signature --
34ca47856c30f97c2a2f21627db7507c4ae661bc

>> FAIL

Topping D50III (4Vrms – Vol=0dB) + A50II (vol at 1 o’clock) – Low Gain, Beyer DT770 Pro, exclusive mode, Filter = mode1.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 17:24:51

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

17:24:51 : Test started.
17:27:05 : Test restarted.
17:27:05 : 00/01
17:27:14 : Test restarted.
17:27:14 : 00/02
17:27:25 : Test restarted.
17:27:25 : 00/03
17:27:43 : Test restarted.
17:27:43 : 00/04
17:27:52 : Test restarted.
17:27:52 : 01/05
17:28:00 : Test restarted.
17:28:00 : 01/06
17:28:12 : Test restarted.
17:28:12 : 02/07
17:28:24 : Test restarted.
17:28:24 : 02/08
17:28:31 : Test restarted.
17:28:31 : 02/09
17:28:44 : Test restarted.
17:28:44 : 03/10
17:28:55 : Test restarted.
17:28:55 : 03/11
17:29:10 : Test restarted.
17:29:10 : 04/12
17:29:59 : Test restarted.
17:29:59 : 04/13
17:31:50 : Test restarted.
17:31:50 : 04/14
17:32:00 : Test restarted.
17:32:00 : 05/15
17:32:31 : Test restarted.
17:32:31 : 05/16
17:32:31 : Test finished.

----------
Total: 5/16
p-value: 0.9616 (96.16%)

-- signature --
aa011171214da04dd145e5209be54a1763a88151

>> FAIL

Topping D50III (4Vrms – Vol=0dB) + A50II (vol at 1 o’clock) TRS-TRS – Low Gain, Beyer DT770 Pro, exclusive mode, Filter = mode3.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 17:37:13

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

17:37:13 : Test started.
17:38:04 : Test restarted.
17:38:04 : 00/01
17:38:41 : Test restarted.
17:38:41 : 01/02
17:38:49 : Test restarted.
17:38:49 : 01/03
17:38:57 : Test restarted.
17:38:57 : 01/04
17:39:06 : Test restarted.
17:39:06 : 01/05
17:39:18 : Test restarted.
17:39:18 : 02/06
17:39:27 : Test restarted.
17:39:27 : 02/07
17:39:37 : Test restarted.
17:39:37 : 02/08
17:39:45 : Test restarted.
17:39:45 : 03/09
17:39:57 : Test restarted.
17:39:57 : 03/10
17:40:14 : Test restarted.
17:40:14 : 04/11
17:40:32 : Test restarted.
17:40:32 : 04/12
17:40:46 : Test restarted.
17:40:46 : 04/13
17:40:58 : Test restarted.
17:40:58 : 05/14
17:41:14 : Test restarted.
17:41:14 : 06/15
17:41:24 : Test restarted.
17:41:24 : 06/16
17:41:24 : Test finished.

----------
Total: 6/16
p-value: 0.8949 (89.49%)

-- signature --
4c508e68b8f1935142a6a79cd1b5a0db0017fceb


>> FAIL

Topping D50III (4Vrms – Vol=0dB) + A50II (vol at 12 o’clock) TRS-TRS – Low Gain, AudioTechnica ATH-M50X, exclusive mode, Filter = mode3.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 18:10:49

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

18:10:49 : Test started.
18:12:10 : Test restarted.
18:12:10 : 00/01
18:12:19 : Test restarted.
18:12:19 : 01/02
18:12:29 : Test restarted.
18:12:29 : 01/03
18:12:47 : Test restarted.
18:12:47 : 01/04
18:12:55 : Test restarted.
18:12:55 : 01/05
18:13:08 : Test restarted.
18:13:08 : 01/06
18:13:21 : Test restarted.
18:13:21 : 02/07
18:13:31 : Test restarted.
18:13:31 : 03/08
18:13:41 : Test restarted.
18:13:41 : 03/09
18:13:49 : Test restarted.
18:13:49 : 04/10
18:13:58 : Test restarted.
18:13:58 : 04/11
18:14:10 : Test restarted.
18:14:10 : 05/12
18:14:22 : Test restarted.
18:14:22 : 06/13
18:14:34 : Test restarted.
18:14:34 : 07/14
18:14:48 : Test restarted.
18:14:48 : 08/15
18:15:11 : Test restarted.
18:15:11 : 08/16
18:15:11 : Test finished.

----------
Total: 8/16
p-value: 0.5982 (59.82%)

-- signature --
c2eaf27cff77efcde0d0b9c4ad307e04e5dd04f3

>> FAIL

Topping D50III (4Vrms – Vol=-3dB) + A50II (vol at 12 o’clock) TRS-TRS – Low Gain, AudioTechnica ATH-M50X, exclusive mode, Filter = mode3.​

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 18:19:08

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

18:19:08 : Test started.
18:19:51 : Test restarted.
18:19:51 : 01/01
18:20:03 : Test restarted.
18:20:03 : 02/02
18:20:12 : Test restarted.
18:20:12 : 03/03
18:21:15 : Test restarted.
18:21:15 : 03/04
18:21:25 : Test restarted.
18:21:25 : 04/05
18:21:35 : Test restarted.
18:21:35 : 04/06
18:21:45 : Test restarted.
18:21:45 : 05/07
18:21:55 : Test restarted.
18:21:55 : 06/08
18:22:05 : Test restarted.
18:22:05 : 07/09
18:22:13 : Test restarted.
18:22:13 : 08/10
18:22:21 : Test restarted.
18:22:21 : 08/11
18:22:28 : Test restarted.
18:22:28 : 09/12
18:22:36 : Test restarted.
18:22:36 : 10/13
18:22:45 : Test restarted.
18:22:45 : 11/14
18:22:57 : Test restarted.
18:22:57 : 12/15
18:23:05 : Test restarted.
18:23:05 : 13/16
18:23:05 : Test finished.

----------
Total: 13/16
p-value: 0.0106 (1.06%)

-- signature --
d36bb27bb9574dba2294dd7737f3559275980c7f


>> SUCCESS (A massive one, actually)

Success, really? Only because the DACs is set to 3dB lower output? Preventing intersample overs, maybe? Nah, pure luck, I randomly clicked, honestly. I should have taken a lottery ticket.

Let's try again the exact same:

foo_abx 2.2.1 report
foobar2000 v2.1.5
2026-08-09 18:25:59

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

Output:
Default : Speakers (TOPPING USB DAC) [exclusive], 32-bit
Crossfading: NO

18:25:59 : Test started.
18:26:08 : Test restarted.
18:26:08 : 00/01
18:26:17 : Test restarted.
18:26:17 : 01/02
18:26:26 : Test restarted.
18:26:26 : 02/03
18:26:33 : Test restarted.
18:26:33 : 02/04
18:26:45 : Test restarted.
18:26:45 : 03/05
18:27:01 : Test restarted.
18:27:01 : 03/06
18:27:07 : Test restarted.
18:27:07 : 03/07
18:27:17 : Test restarted.
18:27:17 : 03/08
18:27:32 : Test restarted.
18:27:32 : 03/09
18:27:42 : Test restarted.
18:27:42 : 04/10
18:27:51 : Test restarted.
18:27:51 : 05/11
18:28:04 : Test restarted.
18:28:04 : 06/12
18:28:16 : Test restarted.
18:28:16 : 07/13
18:28:25 : Test restarted.
18:28:25 : 08/14
18:28:32 : Test restarted.
18:28:32 : 09/15
18:28:43 : Test restarted.
18:28:43 : 09/16
18:28:43 : Test finished.

----------
Total: 9/16
p-value: 0.4018 (40.18%)

-- signature --
0cf4867e22749e0f776acd9c9fc6ceadf6e3cf43

>>FAIL (I should have taken that lottery ticket instead, stupid me)

---

The wife experiment

I like to get my wife involved, she's patient enough with me.

View attachment 550433

After two replies she said "I'm bored, I can't tell a difference. Get me the same with Queen or Tears For Fears, and then we talk".

Tears for Fears +1, Queen -1

 

[pma]

However, since both the converter units and DAC chips are made by the same manufacturers, they might share tech and algorithms, giving them similar conversion performance. It would be great to replicate this test with another unit.

I have posted a couple of abx tests several years ago. Some were not technically good enough, but you might try this one. It is a DAC output vs. CD player output. Cambridge Audio (CS DAC) vs. Vincent (TI BB DAC). So very different DAC devices.

The files should be, and I hope they still are, here:

alphabeta.zip

drive.google.com

Edit: the files are still there, and I am attaching a DW report below. Biggest difference is near 44 s.

 

[pma]

Thank you; this is very helpful. @pma have you considered shorting the input capacitors in the ADC? @IVX would that be a bad idea, even for someone very experienced with working on electronics?

Yes, I did that about a year or two ago. Unfortunately, for recording reasons such "improvement" was unusable. Yes, the low frequency drop and rising low frequency distortion has vanished, but at the expense of huge DC offset in the recorded files and the offset was different in L and R channel. The offset made the Cosmos good only for some kind of measurements, but not as a recording ADC. This DC offset also pushes a DC current to the DUT and the effect depends on output stage coupling of the DUT. So, I have humbly soldered the coupling electrolytes back.

Once again I am returning to this issue. I was able to fix it somehow by this circuit:

The OPA 1622 was able to absorb the huge input DC current of Cosmos that appeared after bypassing the input coupling electrolytic capacitors. The LF distortion and LF drop has gone, as is to be expected.

Improvement of 20Hz distortion was 20dB.

 

[pjug]

Once again I am returning to this issue. I was able to fix it somehow by this circuit:

View attachment 550628

The OPA 1622 was able to absorb the huge input DC current of Cosmos that appeared after bypassing the input coupling electrolytic capacitors. The LF distortion and LF drop has gone, as is to be expected.

Improvement of 20Hz distortion was 20dB.
View attachment 550629

Amazing. Thank you for your great effort.

 

[somebodyelse]

If we get your 10/16, 10/16 and 11/16 results, then we get 31/48 which makes p = 0.02973. Provided that we omitted the 3/16 and 8/16 results. That is why I say it seems that you were able to hear a difference.

What's your reason for excluding the 3/16 and 8/16 results? From the description they're just as valid as the rest, so need to be included.

 

[pedrojoaquimborges]

What's your reason for excluding the 3/16 and 8/16 results? From the description they're just as valid as the rest, so need to be included.

I agree. The fact that I had to insistently focus on just one instrument to the most of my abilities just to discern the tiniest difference illustrates very well how insignificant that particular DAC/ADC conversion is, perceptually speaking. And my job involves doing critical listening for sometimes 6+ hours a day, 5 days a week. I'd imagine the difference is irrelevant/imperceptible for most listeners.

 

[Matt_Holland]

I agree. The fact that I had to insistently focus on just one instrument to the most of my abilities just to discern the tiniest difference illustrates very well how insignificant that particular DAC/ADC conversion is, perceptually speaking. And my job involves doing critical listening for sometimes 6+ hours a day, 5 days a week. I'd imagine the difference is irrelevant/imperceptible for most listeners.

As a highly skilled listener this really just proves the point that the differences (if they are real) are entirely irrelevant. The differences usually described by subjectivists between good measuring DACs would have you believe that they would be obvious to keen listeners. Obvious tonally, rhythmically, and spatially. But the differences people reporting here are analogous to extreme pixel peeping, perhaps even more extreme than that!

 

[pedrojoaquimborges]

As a highly skilled listener this really just proves the point that the differences (if they are real) are entirely irrelevant. The differences usually described by subjectivists between good measuring DACs would have you believe that they would be obvious to keen listeners. Obvious tonally, rhythmically, and spatially. But the differences people reporting here are analogous to extreme pixel peeping, perhaps even more extreme than that!

Oh that's so true. And I don't fully blame them. When, for example, trying to describe the difference between pairs of competently designed, flat-ish loudspeakers, I find that stating that x has more low end than y might evoke a much bigger difference for a reader than what it really represents.

 

[RandomEar]

The differences usually described by subjectivists between good measuring DACs would have you believe that they would be obvious to keen listeners. Obvious tonally, rhythmically, and spatially.

This is a bit of a tangent in this results-focused thread and I assume that you already have a similar view: The problem is simply that the differences those people report are biased results from rubbish testing. Casually listening to one DAC now any then another maybe an hour or even days apart, without level matching or any other proper control gives you absolutely no meaningful impression. All you end up with is some random ideas your brain created independent of any audible clues. So even if audible differences between excellent DACs did exist, they would not be remotely similar to what audiophiles report day in and day out.

As I've said before and as evident from many reports in this thread, proper blind testing is hard work. You need to be concentrated, you need a quiet environment and the right tools. It's neither casual nor easy and - if we're honest - not much fun after the first dozen or so repeats of some small 3 s excerpt played again and again

 

[pma]

Right. And this thread is about finding differences between recorded files, by abx method, and not about sighted impressions. There are hundreds of threads dealing with sighted impressions. AlI - I had my long sighted period as well, of unreliable unsopported results. It ended when I was able to do properly prepared double blind tests. And do not think that if people have measuring abilities and education in electrical engineering and in acoustics means they are deaf and do not listen to music.