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: Newbie Experiencing Strange Phenomena (Read 3593 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Newbie Experiencing Strange Phenomena

I have been experimenting with whether I can hear the difference between Q4 and Q6 Ogg Vorbis encoded tracks using abx testing and I am experiencing something strange.  When I start out with a set of files (I've done it with three different tracks now), I seem to be pretty accurate, getting 7 out of the first 8 or 8 out of the first 9, so I think I can tell the difference, but if I keep testing, my accuracy drops off dramatically, so that I end up at 15 out of 25 or something like that.  Can anyone comment on whether this is common?  What might be causing this?

TIA for your insights

Greg

Newbie Experiencing Strange Phenomena

Reply #1 –
Quote
I have been experimenting with whether I can hear the difference between Q4 and Q6 Ogg Vorbis encoded tracks using abx testing and I am experiencing something strange.  When I start out with a set of files (I've done it with three different tracks now), I seem to be pretty accurate, getting 7 out of the first 8 or 8 out of the first 9, so I think I can tell the difference, but if I keep testing, my accuracy drops off dramatically, so that I end up at 15 out of 25 or something like that.  Can anyone comment on whether this is common?  What might be causing this?

TIA for your insights

Greg

Possibly you are experiencing listener fatigue.  If you think this might be happening, try putting a few seconds of silence in between trials.  This might help, it seems to help me.

ff123

Newbie Experiencing Strange Phenomena

Reply #2 –
The key is to give only answers about which you are certain. When you have a little doubt, stop for a while and go on later.
And yes, it has happened to me when I tried to guess. It is difficult for me to tell if it happens often because I usually use PCabx, which doesn't display the results until the test is over.

Newbie Experiencing Strange Phenomena

Reply #3 –
Is there a minimum number of tests before the result can be considered valid?

Newbie Experiencing Strange Phenomena

Reply #4 –
A casual test is considered as valid when p is inferior or equal to 5 %.
An important or a difficult test is considered as valid when p is inferior or equal to 1 %

But be careful, these values are only valid on the first test, and only when the number of trials that you originally planned to do is exactly reached.

Otherwise, it gives you the freedom to choose among your tests which one has the best p value, that would be cheating.
And if you decide to stop the test once you get a good p value, this is cheating too, because it is nothing else than to choose by yourself among several values given by the program.

More : http://doc.hydrogenaudio.org/wikis/hydrogenaudio/ABX