Yes, of course I will desist from posting on this thread - I only did so in reply to your question about controls - sorry if I was not meant to answer.
Now i understand what your point was. I wasn't clear myself. I had no problem with dropbox download. My problem was with inflexible windows default app to open wav file. Media player is easier but doesn't play wav.
Windows media player plays wav. Did you download the fasteners.zip file and unzipped it then to get 2 wav files? Windows default player also plays wav. I do not know any SW player that is not able to play wav files.
I tried ABX on nail x screw, but no success, I got 6/16. Can't find nothing to catch. I thought that a classical sample with high dynamics and complex instruments spectrum might be easier to tell, but it does not seem so. I am starting to suspect that after removal of -2dB/20kHz drop, it is very difficult to find the difference and that 1% THD occurring at highest amplitudes only is really difficult to tell by listening.
I am attaching a spectrum of the difference file, as a hint where to seek for biggest sound differences, however masking may play a role at high amplitudes.
Did anybody try the new set of sine tones (300Hz, 1kHz, 3kHz)? I find it easy to hear distortion on 1kHz tone, but not on 300Hz and 3kHz tones. Again, in agreement with ear masking curves.
I am attaching a spectrum of the difference file, as a hint where to seek for biggest sound differences, however masking may play a role at high amplitudes.
Did anybody try the new set of sine tones (300Hz, 1kHz, 3kHz)? I find it easy to hear distortion on 1kHz tone, but not on 300Hz and 3kHz tones. Again, in agreement with ear masking curves.
Attachments
I've just tried the sine test. The 300 and 1k were quite easy using laptop speakers and low volume. The 3k one eluded me (a 3/8 and a 5/8) but switching to little earbud headphones and the difference was extremely apparent.
It was these Philips earbuds (they were only freebies with an MP3 player some years ago) that I tried with the Beethoven earlier.
It was these Philips earbuds (they were only freebies with an MP3 player some years ago) that I tried with the Beethoven earlier.
Attachments
Thanks.
Could it be, that you´ve missed my post in the other thread:
Can you tell original file from tube amp record? - test
where i addressed that point (working something out) ?!
Another point that puzzles me is, that it was emphasized numerous times that at the beginning a hypothesis/question has to be stated before starting to work out a test procedure to address it.
Nevertheless you and other are pressing for "some procedure other than the critisized" but noone stated such a hypothesis/question to start.
Is it a matter of recollection, or " i don´t care" or " what jakob is writing about can´t be of relevance" or something else?
I´m (again) sure having emphasized that external validity is of importance beside the internal validity, therefore my warnings that one must be careful to draw conclusions from test results.
More specifically, it´s the reason why i often did question the relevance of differences (multidimensional differences with music as stimulus) that _exclusively_ can be detected by rapid switching approaches.
Hi Karl, thank you, yes I also hear difference between pure and distorted 300Hz tone on notebook speakers easily, but that is for the reason that the notebook speakers have troubles to play 300Hz 😀. With big full range headphones, it is not easy.
I thought this is the ADAD phase related test when i mentioned crescendo sample. For distortion related, vocal is suitable. In real listening the high distortion is actually easier to perceive due to fatigue, but for short ABX it is difficult. For this music i will look for violin that sounds sharper (that's the original).
The 1% thd is a new case for me. Tube second harmonics is fatiguing, and can be used to pick it up in ABX. But if this 1% is clipping, isn't tube soft clipping supposed to be less fatiguing??? I had confusion here because during first listening what i thought an original (sharper) sound became harsher at the last part of the file. Without abx i know which i prefer based on fatigue aspect, but to know which one is the original i will need an abx and look for the sharper note.
3kHz easiest for me, perhaps it's the tiny iem.
Code:
foo_abx 2.0.5 report
foobar2000 v1.3.12
2018-12-16 23:26:52
File A: 300_d.wav
SHA1: cf066fe42a1136f046d21217653d9333afcc800e
File B: 300_o.wav
SHA1: 3e267b09d1c03746379cf8ef4679028b1d9fe17c
Output:
WASAPI (push) : Speakers (Conexant SmartAudio HD), 16-bit
Crossfading: NO
23:26:52 : Test started.
23:28:11 : 01/01
23:28:36 : 02/02
23:28:47 : 03/03
23:28:57 : 04/04
23:29:05 : 05/05
23:29:13 : 06/06
23:29:27 : 07/07
23:29:43 : 08/08
23:29:43 : Test finished.
----------
Total: 8/8
Probability that you were guessing: 0.4%
-- signature --
549c7bdf40a82071f2c86f8b49a1629dd6e609af
foo_abx 2.0.5 report
foobar2000 v1.3.12
2018-12-16 23:32:00
File A: 1000_d.wav
SHA1: 8e7863fb4a59744010a021cae23af1056047fa18
File B: 1000_o.wav
SHA1: 56251f02a845b809ec2dc21bfacdeaf2c62cf84c
Output:
WASAPI (push) : Speakers (Conexant SmartAudio HD), 16-bit
Crossfading: NO
23:32:00 : Test started.
23:32:15 : 01/01
23:32:35 : 02/02
23:32:44 : 03/03
23:32:56 : 04/04
23:33:06 : 05/05
23:33:20 : 06/06
23:33:28 : 07/07
23:33:44 : 08/08
23:33:44 : Test finished.
----------
Total: 8/8
Probability that you were guessing: 0.4%
-- signature --
1e34aefc87b1cbc02b7a72e28e74eeb1d0fd36c4
foo_abx 2.0.5 report
foobar2000 v1.3.12
2018-12-16 23:34:38
File A: 3000_d.wav
SHA1: 05ddaaf4d83064b55e7cabeb05541b251545c9e5
File B: 3000_o.wav
SHA1: 895a995da673366d68fd673ac661184689a5b872
Output:
WASAPI (push) : Speakers (Conexant SmartAudio HD), 16-bit
Crossfading: NO
23:34:38 : Test started.
23:34:51 : 01/01
23:34:59 : 02/02
23:35:16 : 03/03
23:35:26 : 04/04
23:35:38 : 05/05
23:35:56 : 06/06
23:36:09 : 07/07
23:36:17 : 08/08
23:36:17 : Test finished.
----------
Total: 8/8
Probability that you were guessing: 0.4%
-- signature --
5c619db226c77261c8970b5c4ab41e07093e98b2Ah, i think you're right, i remember now. My tablet uses Window8 which i have always avoided in custom installations. I think with WMP i couldn't find the menu to open file? Tried to right-click the wav to find 'open with' or 'properties' but they are not there. I hate windows 8. I'll try again later. May download foobar too.
IOW you are OT here the same as you were there 🙄 Believe me, we all get your message loud and clear
No, clipping of this tube circuit is around 50Vp-p and we play no more than some 12Vp-p.
If you, please, took some short time to read a test description first, you would not think it was a "ADAD phase related test". I am really confused how much of negative impact have the audiophile magazines, discussions and forums to their readers.
Have solved the wmp issue. I don't know how to browse files but music folder is registered as library folder so i just put the wav files in music folder and i could see it from within wmp.
I notice the edited poll isn't allowing multiple choices and there doesn't seem anyway to bring those options back...…
Let me see what I can do.
Hopefully this should be OK now. I'm going to vote
Multiple choices works now so i voted one more time picking my 2 two choices so i again can see poll results up at the top, therefor ask can you into poll remove the one pont choice i did into post 21 "Screw is the original file".
- Status
- Not open for further replies.