Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Foobar testing (Read 4142 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Foobar testing

So I have a friend claiming that he can hear clear audible differences between FLAC and 384 of the same recording.

He tried out Adele blind tests that he claims was not suitable because the album was a crappy master, so he tells me on well recorded material the differences would be easier to detect. One thing he always mentions is listener fatigue with lossy tracks.

So I want to challenge his assertions. So I guess I should ask him to use the most well recorded music he has to clearly hear the difference (to avoid excuses) then use Foobar ABX plugin to test those claims.

My questions are :

1) You can share the results of Foobar with others? So if I want to see how well or poorly he did, I can see those results.
2) Is it possible to tamper with the results from Foobar? Ie acing a test even though he may not have heard a difference?

Just want to make sure when I challenge him to a test of audibility that I have all my ducks in a row, so to speak. Is there anything else I should be aware of to make the test more reliable/valid?

Foobar testing

Reply #1
**Don't show them the results**  - use the *hide* option unless they are experienced and set a good number trials at least 8 and much more for very subltle differences : up to 50

Let him  / her abx for a good while until exhaustion or until they believe they are really winning or losing. Then show results. Why ? Because you can still get good results by just guessing and posting only the good numbers. (but getting <1% is super rare).

If using the sighted option they must get good scores like pval 1% or less. This means  at least 8/8,  9/10, 13/16 etc not the 4/4 BS that has been posted by audiophiles at times.

Foobar testing

Reply #2
Only way to prevent cheating is to be physically present to monitor the test. There is no way to prevent dishonest people from cheating with software.

Foobar testing

Reply #3
1) Yes, you can share.  foobar2000's ABX component creates a text report.
2) It's a text file so yes you can tamper/edit it.

Just let your friend believe what he wants.  My experience is that you will not be successful in changing his mind regardless of the ABX results.

Foobar testing

Reply #4
A signature has been added to the log files in order to address your concern in 2):
http://www.hydrogenaud.io/forums/index.php?showtopic=107354




Foobar testing

Reply #8
Only way to prevent cheating is to be physically present to monitor the test. There is no way to prevent dishonest people from cheating with software.

So true. One can simply route the audio output to an analyzer and know the answer without even listening, assume the differences are visible.

Here is my old method:
http://www.hydrogenaud.io/forums/index.php...st&p=440943

Foobar testing

Reply #9
Only way to prevent cheating is to be physically present to monitor the test. There is no way to prevent dishonest people from cheating with software.

So true. One can simply route the audio output to an analyzer and know the answer without even listening, assume the differences are visible.

Here is my old method:
http://www.hydrogenaud.io/forums/index.php...st&p=440943


Totally agree. History has shown that there is no perfect cure for cheating or stealing. All you can do is raise the bar to success.

Kudos to the Foobar author who substantially raised the bar with the 2.0 ABXplug-in, but regrettably he lives in the same real world as the rest of us...