Yes I did, and I measured. It wasn't the active electronics, it's the headphones. Measuring my Fostex headphones shows they are down almost 10dB at 2K (relative to 1K) and that puts the distortion right at the edge of audibility. So I EQ'd them to a flatter response from 500 Hz to 5kHz. Now I can hear the H2.
See results below. It was easy this time.
Code:
foo_abx 2.0.5 report
foobar2000 v1.4
2018-12-05 17:35:18
File A: sig1.wav
SHA1: 621ecd158928499510d523dc5a85abd019cb1d44
File B: siga.wav
SHA1: 640783db9340fde87db66db7127031862c1d8814
Output:
DS : Speakers (3- BEHRINGER UMC 204HD 192k)
Crossfading: YES
17:35:18 : Test started.
17:35:42 : 01/01
17:35:55 : 02/02
17:36:05 : 03/03
17:36:16 : 04/04
17:36:24 : 05/05
17:36:31 : 06/06
17:36:40 : 07/07
17:36:48 : 08/08
17:36:55 : 09/09
17:37:02 : 10/10
17:37:11 : 11/11
17:37:18 : 12/12
17:37:26 : 13/13
17:37:37 : 14/14
17:37:45 : 15/15
17:37:53 : 16/16
17:37:53 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0%
-- signature --
784fc1580c0aeea16ab525e196062a7ad1602a3bIt was easy this time. With Grado SR80s, I might have got it right off.
This shows that FR of the playback is important in these tests. Now that my headphones are flatter, will I be able to tell the difference in the music files? I doubt it.
That's good. No might about itIt was easy this time. With Grado SR80s, I might have got it right off.
Yes I did, and I measured. It wasn't the active electronics, it's the headphones. Measuring my Fostex headphones shows they are down almost 10dB at 2K (relative to 1K) and that puts the distortion right at the edge of audibility. So I EQ'd them to a flatter response from 500 Hz to 5kHz. Now I can hear the H2.
See results below. It was easy this time.
Code:foo_abx 2.0.5 report foobar2000 v1.4 2018-12-05 17:35:18 File A: sig1.wav SHA1: 621ecd158928499510d523dc5a85abd019cb1d44 File B: siga.wav SHA1: 640783db9340fde87db66db7127031862c1d8814 Output: DS : Speakers (3- BEHRINGER UMC 204HD 192k) Crossfading: YES 17:35:18 : Test started. 17:35:42 : 01/01 17:35:55 : 02/02 17:36:05 : 03/03 17:36:16 : 04/04 17:36:24 : 05/05 17:36:31 : 06/06 17:36:40 : 07/07 17:36:48 : 08/08 17:36:55 : 09/09 17:37:02 : 10/10 17:37:11 : 11/11 17:37:18 : 12/12 17:37:26 : 13/13 17:37:37 : 14/14 17:37:45 : 15/15 17:37:53 : 16/16 17:37:53 : Test finished. ---------- Total: 16/16 Probability that you were guessing: 0.0% -- signature -- 784fc1580c0aeea16ab525e196062a7ad1602a3b
It was easy this time. With Grado SR80s, I might have got it right off.
This shows that FR of the playback is important in these tests. Now that my headphones are flatter, will I be able to tell the difference in the music files? I doubt it.
I see you don't state this but you used a Yamaha HT receiver via HDMI connection on the previous test with Fostex h/p & a Behringer UMC 204HD USB DAC here, presumably with USB with Grado h/p? Are you sure it is the change of h/p that's responsible for the discernible differences?
I did change sound cards, as seen in the log. I can test the Yamaha later. Same Fostex headphones, just with a flatter EQ.
But with the flute files I do hear a difference and will try to prove that via ABX,
I listened to Avocado and Apricot files and could not hear anything to catch hold of, so didn't run a trail. I can't tell the difference even in a sighted test.
But with the flute files I do hear a difference and will try to prove that via ABX,
Sines tested thru HDMI to Yamaha RX-V2700
Code:
foo_abx 2.0.5 report
foobar2000 v1.4
2018-12-05 21:49:29
File A: sig1.wav
SHA1: 621ecd158928499510d523dc5a85abd019cb1d44
File B: siga.wav
SHA1: 640783db9340fde87db66db7127031862c1d8814
Output:
DS : RX-V2700 (Intel(R) Display Audio)
Crossfading: YES
21:49:29 : Test started.
21:49:40 : 01/01
21:49:49 : 02/02
21:49:57 : 03/03
21:50:03 : 04/04
21:50:09 : 05/05
21:50:18 : 06/06
21:50:25 : 07/07
21:50:34 : 08/08
21:50:41 : 09/09
21:50:49 : 10/10
21:50:56 : 11/11
21:51:02 : 12/12
21:51:11 : 13/13
21:51:19 : 14/14
21:51:29 : 15/15
21:51:37 : 16/16
21:51:37 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0%
-- signature --
a9fc471e9b3d38293b709f54815c8a3eb79feb02
Foobar direct x ABX
There were some suggestions that foobar's ABX plug-in is not sending data properly and going through DS, not Wasapi or ASIO. I have to disprove this idea. Earlier I have shown spectrum measurements that were same for direct and ABX playback. Now I took my Cambridge DacMagic+ DAC, which always indicates by front panel LEDs what is the Fs of the file playback. I have chosen first DS as output in foobar settings. Now, with DS output, all the files regardless their format are played at Fs = 96kHz, because 96/24 is my setting in control panel for the DacMagic and windows thus resample to this format. Then I made a change of foobar setting to Wasapi output, and DacMagic LEDs started to indicate properly the Fs of the files according to their format. Then I made a choice of my sig1 and siga_44 test files. Sig1 is in 96/16 and siga_44 is in 44.1/16. The DacMagic front panel LEDs now properly show 96kHz when sig1 is played and 44.1kHz when siga_44 is played. This happens both with foobar direct playback and with ABX plug-in playback. This means, ABX is not using DS (direct sound), but Wasapi. If it was using DS for ABX plug-in, the 96kHz indicator LED would be active when the siga_44 was played. So I consider complaints on foobar ABX playback inappropriate and I repeat they should have been proven, best by measurements.
P.S.: Tried avocado/apricot test through DacMagic+, and though the sound in general is subjectively better to me than with Duo-Capture Ex, I did not get any better ABX result for apricot/avocado test, only 10/16. Sines 16/16 again.
There were some suggestions that foobar's ABX plug-in is not sending data properly and going through DS, not Wasapi or ASIO. I have to disprove this idea. Earlier I have shown spectrum measurements that were same for direct and ABX playback. Now I took my Cambridge DacMagic+ DAC, which always indicates by front panel LEDs what is the Fs of the file playback. I have chosen first DS as output in foobar settings. Now, with DS output, all the files regardless their format are played at Fs = 96kHz, because 96/24 is my setting in control panel for the DacMagic and windows thus resample to this format. Then I made a change of foobar setting to Wasapi output, and DacMagic LEDs started to indicate properly the Fs of the files according to their format. Then I made a choice of my sig1 and siga_44 test files. Sig1 is in 96/16 and siga_44 is in 44.1/16. The DacMagic front panel LEDs now properly show 96kHz when sig1 is played and 44.1kHz when siga_44 is played. This happens both with foobar direct playback and with ABX plug-in playback. This means, ABX is not using DS (direct sound), but Wasapi. If it was using DS for ABX plug-in, the 96kHz indicator LED would be active when the siga_44 was played. So I consider complaints on foobar ABX playback inappropriate and I repeat they should have been proven, best by measurements.
P.S.: Tried avocado/apricot test through DacMagic+, and though the sound in general is subjectively better to me than with Duo-Capture Ex, I did not get any better ABX result for apricot/avocado test, only 10/16. Sines 16/16 again.
Last edited:
Should we deduce from this that you are only interested in flaws in blind listening tests? That would seem to be consistent with many of your posts. It would also be consistent with someone who may have reasons to wish to uphold the value of sighted tests.mmerrill99 said:
I note that once again you have declined to comment on your relationship with an audio website bearing a similar name, and seem to regard the question itself as being some form of "attack". There is a seller of audio cables who happens to have the same brand name as my surname; I am happy to confirm that I have no connection with or knowledge of this business.
Some people do a sighted test and hear a difference; this shows that their equipment is good enough to hear a real difference and they are sufficiently well-trained. They repeat the test blind, using the same equipment. They hear no difference; this is a false negative because their equipment is not good enough and they are insufficiently trained. Is that what we are expected to believe?
I´m not sure why you bring up "sighted listening" in this context. Care to explain?
Generally i´ve written quite often that it is as easy to get incorrect results in "blind tests" as it is with "sighted listening" .
That´s surely true, but did i really dispute that in any of my posts?
Thanks for confirming that Pavel. Be ready for some new reasons why the ABX utility in Foobar is flawed.
Oddly, my install of Foobar does not appear to allow a choice of audio schemes, the only thing in the drop down menu is the names of the soundcards. No DS, ASIO, WASPI. Other audio software on the same computer allow that choice.
Oddly, my install of Foobar does not appear to allow a choice of audio schemes, the only thing in the drop down menu is the names of the soundcards. No DS, ASIO, WASPI. Other audio software on the same computer allow that choice.
One more on apricot/avocado (resampler to 96kHz)
Code:
foo_abx 2.0.2 report
foobar2000 v1.3.7
2018-12-06 13:58:50
File A: apricot.wav
SHA1: 366faaef08134ef83262c95ee8a5ccbdeedb7395
File B: avocado.wav
SHA1: b404fd5526030ace6f56bead6dd1f780cab0a533
Used DSPs:
Resampler (PPHS)
Output:
WASAPI (event) : Speakers (Cambridge Audio USB Audio 1.0), 24-bit
Crossfading: NO
13:58:50 : Test started.
13:59:51 : 01/01
14:00:01 : 01/02
14:00:18 : 02/03
14:00:26 : 03/04
14:00:36 : 04/05
14:00:45 : 05/06
14:00:56 : 05/07
14:01:05 : 06/08
14:01:27 : 07/09
14:01:42 : 07/10
14:02:01 : 07/11
14:02:18 : 08/12
14:02:29 : 09/13
14:02:40 : 09/14
14:02:50 : 10/15
14:03:05 : 11/16
14:03:05 : Test finished.
----------
Total: 11/16
Probability that you were guessing: 10.5%
-- signature --
ddd3722f50ecd8bd0ef3f809479a09ab4706903bThanks for confirming that Pavel. Be ready for some new reasons why the ABX utility in Foobar is flawed.
Oddly, my install of Foobar does not appear to allow a choice of audio schemes, the only thing in the drop down menu is the names of the soundcards. No DS, ASIO, WASPI. Other audio software on the same computer allow that choice.
Wasapi plug-in for foobar must be downloaded
foobar2000: Components Repository - WASAPI output support
and installed to foobar. The sound card must enable wasapi and have the proper driver. Same applies to ASIO. Basic installation is usually only windows direct sound. Maybe the receiver even does not have an option of ASIO or WASAPI drivers.
What is WASAPI?
If your card does not support WASAPI or ASIO, please at least disable all enhancements in control panel. Otherwise windows would create dirty spectra.
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.