is a very famous post from Head-Fi, which is the most detailed blind listening test summary of various metaphysics in hifi that I can find. Of course, as the original author said, some of them are not ABX tests, but just blindly listening to reviews. I am not responsible for some of the test results in this sect, but just translated the original post. The original text is not very smooth, I have tried my best.
Testing audiophile claims and mythswww.head-fi.org/threads/testing-audiophile-claims-and-myths.486598/
Translation in spare time is inevitable. Here is the original post:
(I will add them to the list here as I find more blind listening test results.)
We like to have a good discussion/argument/general about the metaphysics that enthusiasts have done that many others think are nonsense. These arguments have been going on: I hear different sounds—but it is impossible to have any difference, it is all the difference between the sounds you imagined—have you tried different fever wires?—I don't need it to be in your mind, and so on, we all know how it works.
Occasionally someone tries to verify these myths. For example, three What HiFi forum members were invited to their listening studio and had passed a series of blind listening tests from fever cables to sound source code rates. From the questions I have read, it is certain that the statement that "difference is metaphysics" is incorrect, and the difference is real. Different bit rates have been correctly identified, and different fever wires produce different sounds in the same HiFi system. However, these are just blind listening and reviews, which is different from the ABX test that requires people to correctly distinguish products.
Here are some blind listening tests and ABX test results I found online. All I did was to sum up their conclusions.
What you need to pay attention to is the difference between blind listening and ABX tests, because they will produce different results.
blind listening test refers to the listener who does not know what he is listening and is asked to describe the differences they hear. This is a commonly used blind listening evaluation method in audio. The results of such tests are often "surprising" performance of low-priced high-fidelity, high-priced factors, appearance, product reputation and other factors that listeners do not know. Some blind tests also involve competition between products, such as the PK of two amplifiers, and the winner enters the next round. As you can see, my definition of blind listening tests is very broad.
ABX test is more of a test. You listen to Product A and Product B and then play X, is it A or B, and you have to say what it is. There may be more than A and B, as some tests involve multiple fever wires. Then any difference must be clearly expressed, and this is not the case for equipment similar to fever wires. My definition of ABX testing is also very broad.
aims to see what the overall results of these tests are and whether they provide evidence to support or deny the so-called old-shao metaphysicism. Before you read the test below, you can try it yourself to see how your listening is and understand what ABX test is...
- ABX double-blind listening comparator
This is a website dedicated to such testing. In May 1977, people compared the amplifier. It should be noted that not all ABX tests are negative. Some people did find the difference. This suggests that some parts of the HiFi chain do differ, but others do not.
A test of speaker cables found that no one could find the difference between speaker cables from $2.50 to $990. All the results are even close to 50%
There is an interesting comparison to finding that 'video wire', once it exceeds 50 feet (about 15.2 meters), compared to 6 feet long (about 1.8 meters) wire is easily found.
DAC is not suitable for CDP found that original CDP can be distinguished from more modern CDP, but expensive standalone DACs are the same as CDP.
None of the tests involve a large number of people, some are just one person.
DACs don’t fair well with CDPs finding an original CDP being distinguishable from a more modern one, but an expensive stand alone DAC being the same as a CDP. (I really don’t know how to translate this sentence)
These tests did not involve a large number of people, and some experiments were just done by one person themselves.
. The influence of the interaction of wire, speaker and amplifier, 1991 International Audio Engineering Paper
A total of 12 wires were tested from Levinson to Kimber, including car jumper and light wire, with prices ranging from US$2/meter to US$419/meter. The result is based on the theory that the speaker cable should transmit all frequencies from any amplifier to any speaker without loss, and the loss is due to resistance. There is a hypothesis that making more frequencies and less distortion will sound better. This seems reasonable in my opinion.
The best performance is multi-core wire. Car takeoff leads are not done well, and cables for digital transmission are done well! The most expensive wire is not mentioned in the conclusion, but the cheapest wire is praised for its performance, and Kimber does it well. Unfortunately, there is no clear list of wire costs and performance, so it is not clear whether cost equals performance, but it is recommended that structure equals performance.
. Do all amplifiers sound the same? The original Stereo Review blind listening test
tested many amplifiers of different price ranges and types. Listeners claim to be believers and skeptics about whether the audiophiles’ statement is true.
has 13 links, and each time there are different numbers of listeners. The performance of skeptics and religious believers is not much different. The 2 skeptics get the highest correct score, and 1 get the lowest correct score. The overall average is 50.5%, which is the same as the random guess. The cheapest Pioneer amplifier is perfectly capable of over the more expensive amplifier, and it is "amazingly similar to Levinson".
Apart from that, to explain why the amplifiers sound the same, here is an Wikipedia entry for Bob Carver and his amplifier blind listening challenge.
. Wire Directionality
is not the best link because it only involves one test and does not give much details. Fever wire maker Belden conducted a test with a less-known magazine and found that the results were completely random.
I like the next sentence, "Belden is still happy to manufacture and sell targeted wires to enthusiasts"
. Head-Fi wire ABX test, August 2006
From Canare, Radio Shack and Sliver three fever wires are placed in the same snakeskin net and are disguised. Each has a mark on it, so that only the person who initiated the test knows which one is which one, and then sends it to different forum members. As a result, only one forum member answered all three questions correctly. Radio Shack's cheap cables and Sliver's are the hardest to distinguish.
Unfortunately, I can't see from the article how many members are participating and what the exact result is.
. HiFi Wigwam Power Cable ABX Test, October 2005
This is a good large-scale ABX test. Similar to Head-fi, forum members also received four wires, including 2 electric kettle cables (a backup power cord included with the HiFi product), an enthusiast cable, a DIY cable and a test disc. These results are uncertain to say the least;
electric kettle line is C. There are 23 answers:
say the kettle line is A
say it is B
say it is C
say they don't know.
The general conclusion is that the electric kettle cable is not correctly identified, or that one power cable is better than the other.
A participant in this test pointed out that the two kettle leads described as exactly the same in the test are actually not the same, just the basic leads of the HiFi product.
. What HiFi The biggest question about wires 2009
started in September 2009. Three forum members were invited to What HiFi (WHF) for a blind listening test, and they thought the toolkit (Roksan, Cyrus, Spendor) was changing, but instead, the cable was. The same three tracks run through the whole time.
This test started with the cheapest wire you can find with What HiFi, no one liked it and claimed it sounded bland and boring. Then a Lindy power regulator and Copperline Alpha fever power cord were introduced, and the sound was improved.
audio cable was changed to Atlas Equators, and two of the 3 tracks are said to have improved, with both low frequencies and details becoming better.
Finally, the speaker cable of 60p /meter is changed to Chord Carnival Sliverscreen at 6 pounds /meter. Similarly, people have noticed changes, but not much.
has undergone various exchanges after that, which confirms the above situation, namely the power cord has the greatest impact. When the test results were revealed, the participants were very surprised!
However, this is not an ABX test, but a blind listening review. When you read the initial explanation of this article, you will find that these two tests will produce different results. Worryingly, when I asked Chief Editor-in-Chief Clare Newsome about such tests, she claimed that the tests were ABX, while elsewhere on the forum, they claimed to do ABX tests. But they didn't, they were blindly listening to the reviews, which gave people the opportunity to claim differently, but without providing evidence, they could really hear differently.
. Home theater and high fidelity secret. Can we hear the difference between different power cords? A Blind ABX listening test. In 2004,
, a comprehensive article with pictures, in 149 tests, the total result was 73, so the accuracy rate was 49%, which was no different from randomness.
. Boston Audio Association, an ABX blind listening test against Ivor Tiefenbrun, founder of Linn, a rather complex test for Ivor Tiefenbrun himself in 1984. He was very supportive of vinyl record and against digital audio at the time (almost the opposite of Linn's current operation!). There are various tests, and the general conclusion is:
"In short, in this series of tests, Tiefenbrun provided no evidence to be able to reliably identify:
(a) There is an undriven unit in the room,
(b) The Sony PCM-F1 digital processor appears in the audio chain, or
(c) The relay contact of the A/B/X switch box exists in the circuit. ”
Even the founder of Linn cannot support the statement he made when he was tested by ABX.
0. Famous Audioholics forum post, wire and clothes rack , June 2004
https://forums.audioholics.com/forums/threads/lfe-problem-in-marantz-receiver-sr-4000.24/#post-97
1. Spain's Matrixhifi.com forum, ABX tests on two systems, June 2006.
Two systems, one is a cheap (A) Sony DVD player and Berling amp (supported on the folding chair), with chepo cables; the other is a more expensive (B) with Classe, YBA, Wadia and expensive fever equipment and a suitable rack, hidden behind the scenes and connected to the same set of speakers.
The result is:
8 people participated in this test
4 people chose the "A" system as the best system
0 people chose the "B" system as the best system
4 people could not hear the difference, or did not choose any one as the best.
2. AVReview, Wire Blind Listening Test, May 2008.Some members of the
AVR forum went to a HiFi store in Sevenoaks and listened to the same set of devices with two cheap Maplins cables (priced £2 and £8, respectively) and a Chord Signature (priced £500). They found that cheaper Maplins wires are easy to distinguish, and more expensive Maplins are harder to distinguish from Chord. Their residents skeptics agreed that he could hear the differences. The final conclusion is;
... From our nearly 20 test samples, we got 14 correct answers. The result is 70% correct. . . .
This is the second ABX test that shows that the cable does differ from what HiFi. But as What HiFi shows, the difference between blind listening tests and ABX test results, and the ease of trying and blurring both tests.
3. International Audio Engineering Association AES, ABX test of CD/SACD/DVD Audio, September 2007.
One of the summary states that "a carefully controlled double-blind listening test attended by many experienced listeners, and the experimental results show that no difference between formats can be heard". The result was that in 554 trials, 60 listeners could not hear any difference between CD, SACD and 96/24.
Note that there are flaws in this experiment, basically the high-resolution sound source is converted from the original CD.
4. What HiFi, HDMI cable blind test, July 2010
Another What HiFi test, the changes that the three forum members are making are using three HDMI cables. To their knowledge, the device may be replaced. The wires are free, £75 Chord and £150 QED. Throughout the test, all three struggled to find the difference, but more confidently, the difference was in the sound rather than the picture. They prefer free cables over chord cables and find it as good as the most expensive QEDs. This result is common in blind trials and really distinguishes it from the ABX trials.
In my opinion, these reported differences can be explained by the three brave testers saying that there is no difference. They were invited to take a test, hoping to find out the difference.
5. Harman International (AKG, Infinity, JBL) Floyd Tool; Audio, a science serving the arts, 1998.
A paper written by Floyd Toole deals with many topics of scientific measurement and audio. Go to pages 10 and 11, and there is a paragraph about blind listening tests. It shows how the “difference” between speakers is greater when performing non-blind listening tests rather than blind listening tests. The obvious conclusion is that non-blind listening tests can cause factors other than sound to play a role in determining what sound is better.
6. Sean Olive, Director of Acoustics Research, Harman International, Disgrace in Non-blind Listening Test , 2009.
Study on 40 Harman employees and compared the results of blind listening and non-blind listening tests of four speakers. As Director Harman’s colleagues said, auditory tests that allow the use of visual displays biases that are not found in blind listening tests.
Below the article are various responses to the blog, including a very interesting exchange between Alan Sircom, editor of Hifi Plus magazine and Sean Olive. Alan Sircom makes the very interesting point that volume has a role to play with blind tests
"Here's an interesting test to explain what I mean: run a blind test a group of products under level-matched conditions. Then run the same test (still blind), allowing the users to set the volume to their own personal taste for each loudspeaker under test. From my (admittedly dated and anecdotal) testing on this, the level-matched group will go for the one with the flattest frequency response, as will those who turn the volume 'down', but those who turn the dial the other way often choose the loudspeaker with the biggest peak at around 1kHz, saying how 'dynamic' it sounds."
I had not thought of that before. You will end up with different conclusions between a blind test where the volume is set and where the volume can be adjusted. Adjustment allows preferenCES for different sounds to be expressed, without other influences being present that clearly have nothing to do with sound.
7. Russ Andrews reconnected David Gilmour's recording studio in 2000.
This is not a blind listening test, but I think it is worth putting it in this post. The studio used by David Gilmour (I think he owns) was reconnected by Russ Andrews using a Kimber cable. This is obviously tested in extensive AB comparisons. I'll love the ABX test later!
8. DIY Audio Forum, a player's confession, 2003.
A forum member joined in and admitted that "then I started hearing about some compelling blind listening tests, and finally I did the test myself. I was shocked by the results. In the store where I worked, there was no difference between a $300 amplifier and a $3000 amplifier. No one else working there could be." Then he did a blind listening test on a friend himself, which was an Onkyo SR500 Dolby Digital amplifier and a Bryston 4B 300 wpc amplifier + a Bryston 2-channel preamp - this is his friend's system. Friends with "blush faces" can't tell the difference.
9. Boston Audio Association, Discussion and Analysis of Two Blind Listening Tests, 1990.
In an article discussing the Stereophile CD-tweek blind listening test, "In the CD-tweek blind test, Atkinson and Hammond conducted 3,222 single-blind listening tests to determine whether CD-tweak (green ink, armor, expensive transport) changed the sound of CD playback. Overall, subjects could only identify adjusted CDs and undefeated CDs 48.3% of the time, and the proportion of high scores (five, six or seven out of seven trials - the definition of a keen listener by stereo enthusiasts) was completely within the expected range when the subject was merely guessing."
0. Cowan Audio, an Australian enthusiast, blind listening test between different CD machines, 1997.
An unknown model of $1,800 (they don't want to disclose the model) compared with a Sony CD machine with a $300 Sony CD machine, and the result was only about 50% accuracy, which is no different from being blind. William Cowan first performed a non-blind listening test and claimed that “it’s too easy, let’s continue with the blind listening test.” oops!
1. Pio2001's own ABX test between CD and vinyl in Hydrogenaudio 2003
The results were 3/7 and 5/8 correct.
2. Tom Nousaine, article to Tweak or not to tweak? 1988.
A test using the same CD and speaker, but with different amplifiers and cables. One is $300 and the other is $5,000. There were 7 listeners, and 10 trials failed.
3. AV Science Forum, Monster VS Opus MM 2002
is not particularly strict because there are not enough tests, but as the poster says, "And to get to the point, Mike cannot accurately identify Monster and Opus MM in our test method (or, in turn, if he is always wrong, this will be a positive result). We stopped testing less than halfway, and I think we went through 8 A/B before giving up."
4. http://Stereo.net, a blind listening test of two preamps, May 2008.
This is an Australian forum, so the conclusion is usually straightforward "Conclusion: There are errors between the two preamps, they are very close and no difference can be reliably picked." The tests run well despite the testers seemingly have questions.
5. Stereomojo, digital amplifier comparison test, in 2007,
multiple amplifiers were conducted for blind listening test, one group of two PK, and the better performing amplifiers passed the test and entered the next round. The winner is a cheap amplifier called the Trends Audio TA-10, which costs $130, and it is a small amplifier in the upper right corner of the stack.
6. Head-Fi Wire ABX Test 2006
Three fever wires are Canare, Solid Silver and Rat Shack, but after disguising, the appearance looks the same. Only one person can tell the difference, and when there is no audible difference, people are likely to be guessing, you will expect the difference to happen.
7. A blind listening test for HDMI cables in a French website
Nine participants used nameless wires, Belkin and Magic Sound HDMI cables. Only one person claims to be more inclined to which line, but his feedback is inconsistent with his claims.
8. Home Cinema Fr.Com A French test 2005
These wires include Taralabs, VDH, Audioquest (American Line Saint) and a DIY wire. The result is that no one can reliably tell the difference.
9. Sound & Vision. Tom Nousaine Article on 3-time speaker cable blind listening test
All three tests failed, and the listeners used their own equipment, music, volume and time.
Note - Unfortunately Tom Nussen passed away, the direct link to his article has been interrupted.
0. Crazy sound, blind listening test CD with digital files, expensive speaker cable with cheap cable. Wall Street Journal " January 2008
tests at an audio-visual exhibition in Las Vegas found that when compared with CD, Wav files (52%) were better than MP3 (33%). In a comparison test of Sigma speaker cables and hardware cables worth $2,000, 61% of the 39 people participating in the test prefer more expensive cables. There are no substantive conclusions about these test results, but interestingly, John Atkinson and Michael Fremer from Stereophile magazine are described as being easy to pick out more expensive wires.
1. AV Sience Forum, Observation Results on Audio Test of Controlled Wires November 2007 Blind listening test between
Monster Wires and Opus MM, as far as I know, this is a wire worth $33,000.
However, the owner of very high End equipment and wires cannot distinguish the difference between the two.
2. Audio Critic, ABX test on amplifiers, 1997.
Tom Nousaine wrote to an audio critic in which he described the ABX test of a very High End system owner, in which the Pass Labs Aleph 1.2 200w mono amplifier was randomly replaced with the Yamaha AX-700 100w combined amplifier. In the first test, the High End equipment owner identified 3 of 10 times, followed by 5 of 10 times.His wife was right 9 times 16 times, and his friend was right 4 of 10 times.
3. Expert Reviews. HDMI cable blind test, February 2011
two TVs, two Sony PS3 and a James Bond movie played side by side. The only difference is to replace HDMI cable. Interestingly, the difference in this photo is small, but the difference from the sound is more obvious. However, many people prefer cheaper sounds compared to expensive cables.
Note- not ABX test, and commenters admit that there may be slight differences between TV and PS3.
4. Stereomojo, blind listening test for 6 DACs
Contrary to another blind listening test, the ABX test found that there is only a slight difference between the cheapest and most expensive DACs.
5. Wilson's iPod experiment, CES 2004. Stereophile 2004
Paragraph 10. A "trick" blind listening test, at the CES show, a group thought they were listening to a $20,000 CD, but in fact, they were listening to an iPod.
Vision will really play a big role in sound.
6. A whole night of listening to different wires, AVForums June 2006
In addition to the above ipod experiment, a member of AVForums reported his experience of performing non-blind listening tests and blind listening tests at a dealer. Conclusion of the comparison test of
"This is what I heard.
- With one exception, all wires sound subtle differences.
. Some music differences are not obvious
. My evaluation when I listen blindly is different from my evaluation of "non-blind listening"
7. A blind listening test for classic and modern violin , Westerlunds Violinverkstand March 2006
This is really interesting, but it again shows how different we are when we are doing non-blind listening tests and blind listening tests. In this test, six violins, three c1700 (including one Stradivari) and three modern violins played to a group of string teachers who voted on their favorite violins. The stage was dark and they couldn’t see which one was. Stradivari ranked last, and a modern brand won.
8. The edge of audibility, blind tests on recordings with or without filter . June 2011 Pink Fish Media forum
You can download and try recording yourself. Among those who have already, 2 choose No. 1, others choose No. 6, and no one choose No. 10.
9. Try the bit rate blind listening test
A very good setting and easy to use test for different bit rate blind listening websites.
0. Blind listening test for CD turntables, StereoNET Australia - Hi Fi & Home Cinema News October 2008
is a good setup and description. After a set of failed tests, they admitted that no one could hear the difference, gave up, but drank some beer!
1. ABX test with different levels of jitter jitter added to the track, HDD Audio forum March and April 2009
A member recorded his score, which was no different from random.
2. Stereophile Regarding the ABX test of amplifiers, 1997
total of 505 listeners participated and came up with the following carefully crafted results chart
This is a random bell-shaped curve, as you can guess. However, Stereophile claims that the test was successful because some people did better than average. There may be some facts because there have been blind listening amplifier tests that passed. Even so, this small part of the subjects need to test again to confirm whether they were due to luck. The statistical significance of this test is not enough to indicate that there are audible differences.
3. Head-Fi, a member conducted non-blind listening tests and blind listening tests on the wires, November 2011.
This provides more evidence that vision tests and blind tests produce different results, i.e. people can hear the difference when they can see the product, but cannot hear the difference in blind listening tests.
4. Minnesota Audio Association. Wire listening test. April 2012
The results are very complicated, and no wire can make obvious differences. They acknowledged that there was no objective difference, but because they found a difference that was easily explained by random selection, they concluded that there was a subjective difference, and so-called "wires do have an impact."
5. Richard Clark Amplifier Challenge, June 2006
"Richard Clark Amplifier Challenge is a hearing test designed to show that as long as modern power is working within its linear range (below the limit), the human ear cannot hear the difference between the amplifiers."
This is an ABX test, which requires two sets of correct identification 12 times. More than a thousand people were said to have participated in the challenge, but none of them passed.
results show that I should buy the cheapest amplifier?
No, you should buy the best amplifier. Factors to be considered are: reliability, quality, cooling performance, flexibility, mechanical connection quality, manufacturer's reputation, special characteristics, size, weight, aesthetics and cost. Purchasing the cheapest amplifier may give you an unreliable amplifier and is difficult to use and may not have the required features yet. This test shows that the only factor you can ignore is the sound quality that is less than clipping. ”
This is a relief for those who spend a lot of money to buy a beautiful amplifier.
6. Audio Video Revolution Forum, blind listening test for speakers, November 2007
Good news is that the results strongly show that even under blind listening conditions, there will be obvious differences between speakers, whether subjective listening or objective test.
7. PSB speaker, a comparison blind listening test of four speakers, November 2005
The author is very happy that he did not choose the cheap speaker, but he did not mention whether the speaker is easily identified.
8. Which? Consumermagazine, 2016
"UK consumer magazine" (UK consumer magazine) values its independence and testing procedures to provide fair and independent advice.
" tests show that cheap HDMI wires—even wires worth only a few pounds—can work as much as more expensive wires. When we last performed the HDMI test, we found that there was no significant difference in image quality between an HDMI cable worth £10 in the supermarket and a cable worth nearly £100. ”
9. Believe me, I am a scientist - Audio Poll: Neil Young and High Resolution Audio, May 2012
Comparative test of high-resolution WAV version file of the first album of the same name vs. standard AAC file.
"Most of you are audio engineers, professional musicians and ambitious enthusiasts, I think if anyone can distinguish these file types, it's you.
So, what is the result?
OK... Please accept my warm congratulations, congratulations to you 49% of you for guessing it right.
Yes: Even among our readers, the result is not much better than tossing a coin. We don’t even need a large sample size to get results consistent with the vast number of studies already done in this field."
0. Analogue Planet, evidence of huge sound differences in fever power cords, January 2018
" However, comparing the "stock" rubber crust wires and AudioQuest fever wires, the most important thing is to prove that there is a difference in the power cord, and you should be able to hear it even with a second row of gun microphones! "
watch the video, and it starts about 33 minutes, the sales promotion is over, and then play the same music with different power cords. Can you hear different sounds?
Conclusion
The clear conclusion is that the ABX test results do not support the statements of many enthusiasts, so they become the metaphysics of enthusiasts because they show that the wire itself does not change the sound. Any changes in sound quality come from the listener's imagination and senses Interactions. The difference claimed to be heard is not reliable. Blind listening tests are sometimes regarded as ABX. But Blind listening tests are not really ABX tests, it is a test of a product without seeing it and allows for an unverified sound to be declared.
If HiFi is about sound, more specifically about sound quality, then even if other senses are removed, we should still be able to hear differences, and these differences can be verified by listening to identify one product and another. But I We can't hear these differences, failing again and again.
Therefore, you can buy good but inexpensive HiFi products such as cables, amplifiers, CDs, etc., and be content with the sound they produce is excellent. You do need to spend time choosing speakers because they do sound obvious differences. Or you can buy expensive HiFi products, such as fever cables, etc., imagine and distinguish between one hifi equipment and another hifi equipment by appearance and sound. But you can't just use the sound to buy Expensive equipment is distinguished from cheap equipment. But it is important.
https://www.stereophile.com/?ifaud=true#wsTQZ0dOJYDcJv5K.97
A Stephile employee said after a failed blind listening test:
"For example, more than 10 years ago, during a blind test organized by Martin Colomers, I failed to put Quad The 405 is distinguished from the Naim NAP250 or TVA tube amplifier. I believed in the validity of these results, consistent with what the Consumer Reports said, and I sold my exotic and expensive but made me happy Levinson amplifier and bought a much cheaper Quad 405, the biggest mistake of my career! "
The author of the article continued;
" My point is that taking any kind of blind listening test will inevitably cause an unnatural state, a situation we have never encountered before when listening to music and feeling happiness. ”
I 100% agree, I did the same tests, read and research these posts, I will connect my MF XúCANV8P head amp from using my Firestone Fubar DAC and power supply. All the test status sounds the same. But they don't!!! That's because when I listen happily, I can see my device, its red, green and blue lights on telling me that it is working. This makes me happy, and happiness makes the sound quality better.
above is all the content of the original post's main text, and a reply below makes me speechless:
For some enthusiasts, this is actually not related to "love the sound" and "believe in your ears", but is about having the most expensive sound system, knowing the least about it, and having a long-term hobby and shameless claims about knowing what sounds best. I think everyone needs to feel important and experience the midlife crisis in their own way (some people will never end).
Abx tests are actually introduced in my article two days ago.
ABX test is a way to compare two sensory stimuli to determine if there is a detectable difference between them.First, two samples A and B are played separately, followed by an unknown sample X. X is randomly selected between A or B. Participants need to give clear judgments on whether X is A or B. If X cannot reliably judge by a low hypothesis value (p-value), it cannot be proved that there is a perceived difference between A and B.
ABX test can be easily performed in the form of a double-blind listening test, eliminating any potential impact from the researcher or test organizer. Since Samples A and Sample B are provided before Sample X, there is no need to distinguish the differences between them based on assumptions of long-term memory or past experience. Therefore, the ABX test answers whether there is a perceived difference in ideal situations.
ABX test is usually used for evaluation of digital audio data compression methods; Sample A is usually an uncompressed sample, and Sample B is a compressed version of A. Audio compression artifacts that indicate defects in compression algorithms can be identified through subsequent tests. The ABX test can also be used to compare the fidelity loss levels of two different audio formats at a given bit rate.
ABX tests can be used for audio input, processing, output components and wiring: almost any audio product or prototype design.
If only one ABX trial is performed, random guessing there is a 50% chance to choose the correct answer, just like to tossing a coin. To make the statement have a certain degree of credibility, multiple trials must be performed. By increasing the number of trials, the likelihood of statistically determining a person's ability to differentiate between A and B increases at a given confidence level. A 95% confidence level is generally considered statistically significant. QSC recommends that at least 10 hearing tests be performed in each test round.
5% confidence level results required
5% confidence level results required, i.e. if 10 tests are performed, then you should answer at least 9 times correctly, and so on.
usually think that the results of 16 tests are more convincing. However, some people think that more times can be performed, and the test results can be analyzed through grouping, and the people participating in the test can be given sufficient rest during the period.
comes with a common music playback software Foobar's official ABX test plug-in. You can try the difference between 320k MP3 and WAV. I believe that many people will doubt their life~
https://www.foobar2000.org/components/view/foo_abx
ABX is a forced selection test. A selection of an experimental subject can be valuable, i.e. the experimental subject does honestly try to determine whether X appears closer to A or B. If one person finds no difference in multiple tests, this may dilute the results of other subjects who are focused on taking the test, making the results consistent with the Simpsons paradox, resulting in a false summary result. Simply looking at the total number of test results (m of n answers are correct) does not necessarily reveal the occurrence of this problem. (That is, the results of individuals need to be analyzed one by one)
Strictly speaking, ABX test is a type of double-blind listening test. But in reality, the ABX test is used to determine whether the difference can be heard, and the double-blind listening test (such as MUSHRA) is used to evaluate which subjectively is better when different differences can be heard, and how big the actual difference is between the products, and to exclude other bias factors other than sound interference.
If the difference is small, people participating in the test may feel frustrated and simply complete the test by random voting. So when the difference is small, forced selection tests like ABX may tend toward negative results if appropriate measures are not taken.
However, I personally think that for the "one ear difference" and "heavenly different" that some people claim, they should not have this kind of frustration.
The human senses are not as reliable as imagined, nor as sharp as imagined. We can tell the red and green, we can tell the ups and downs. But the ability claimed by some audio enthusiasts and wine tasters is a bit too extreme. If they are really talented, they should be tested and proved to everyone.
——A netizen in Quora forum
PS: I don’t understand wine, but I have tried blindly testing cappuccino and latte by myself, and I can still tell it.
Finally, I sighed that the rise of Star Wars 9 Sky Walker is indeed a failure, but I have recently revisited the first six Star Wars movies, which are indeed so classic and very good. In addition, the seventh part of the Force Awakening is OK. After writing this article on ABX test, I suddenly remembered a famous saying by Master Yoda:
Do or do not, there is no try~
If you really click the link of Foobar ABX comparison plug-in above and test it, I hope the Force will be with you~
(By the way, update the Sony A105 player review tomorrow)
The big wets are making trouble again and say that abx blind listening is not scientific because the listening time is too short and it is impossible to tell.
But the big wet people usually have one ear, why can’t abx have one ear?
In fact, those who have a little understanding of abx or who have done abx know that this statement is pure nonsense, or they have never done blind listening at all. Because abx does not limit the time at all, listen as long as you want. I can listen to it for a month.
Before, fraudsters liked to say that blind listening time is too short and they are not familiar with it. However, anyone who has done abx blind listening test knows that this is pure nonsense. Because abx double-blind listening does not limit the time , listen as long as you want. There is no limit to the sound source, you can choose music you think is familiar with. The big wets saw that this fraud statement was really untenable.
Now the big wets say that the middle switching interval is too long, affecting their memory.
This shows that although the big wets have been cheating for so long, they are not right to say blindly listen, but they have never tried blindly listen.
Because I have done the foobar abx test, I know that the switching is instant. And you can even choose whether to switch to continue playing or play from scratch. If someone thinks that the whole piece of music is too long and makes excuses, there is actually no mandatory time limit for testing music. In theory, you can also choose a shorter edited music clip. abx does not limit the number of switches. You can listen all the time if you want to listen.
Entities like earphone player can use physical switchers like well-known directors. As long as there is no relay for the physical switch, the switching can be completed in an instant.
We assume that this time interval is 0.1 seconds. Yes, the heavy humidity is only 0.1 second.
When you listen to the big wet, how many seconds do you change the equipment? How many seconds does it take to change music? It probably won't be so fast, right? How can I remember it normally?
But since Dashi himself can't remember what he said, it means that Dashi only has 0.1 second, and everyone here is a witness~
And as mentioned before, abx does not limit the number of switching times, and can theoretically switch it all the time before reaching the conclusion. We assume that the first time wet really doesn't work, and we can't hold on in 0.1 second. So what about twice? What about three times? If the result is still wrong when switching 1,000 times, does it mean that heavy wetness is really not good? But why is it different to have one ear when Dashi only changes the equipment once?
abx double-blind listening can effectively expose the cable ear playback player and other scams. I hope everyone can spread this method so that more people can avoid being deceived and spend meaningless money.
Vitality? Stop funny. Audio as a hobby is dying, largely because of the audio itself. In the real world, Hi-End speakers have lost their reputation in the 1980s because they categorically refused to use basic impartial controls (such as double-blind listening tests). This rejection has become the source of the endless ridicule and ridicule of rational people, and this is also the source of my infinite embarrassment, because it is my supporters who have caused such a chaotic situation, and I am to blame. For the sake of statement: I have never said that the measurement results are not important. I'm saying (and often say that), they don't always tell the whole story. Not exactly the same.J. Gordon Holt, founder and CEO of Stereophile magazine, said in an interview in 2007. J. Gordon Holt died of illness in 2009.
