What never gets explained about ABX testing.

[pyrrhon]

It is all too familiar experience to me at least that you spot your ears/brain has done some work, wihropus asking your permission!

What you say I think is true. At the same time we have this what we want to think as our intellectual part odf us. That is actually very smart. E.g. I can now revoke in my memory a set of different sorts of understandings & emotions evoked by the charateristics of the sound in a particular system produses and then make generalisations concerning the "brains memory music signal analysed results" and learn what good sound actually is. I think this is what amplifier & speaker designers are able to do & do as their work!

You are the special case here cause I knew upfront you would understand me! Look we have some crazy listening sessions ahead with all that knowledge they GIVE us. I trully hope to see you back around in the future.

 

[pyrrhon]

It is all too familiar experience to me at least that you spot your ears/brain has done some work, wihropus asking your permission!

This one just for us : blues the healer !!!!!!!!!!

 

[steve_1979]

This one just for us : blues the healer !!!!!!!!!!

https://www.youtube.com/watch?v=x0isR1emmPs

 

[steve_1979]

I have seen some ABX test result but never in the scope of days where people have been accustomed to one system before a change got introduced. Well my bad here I should have checked those tests methodology. The ones I heard of are always rather quick switching and the user have to guess the hifi vs the cheap one. Like I said if my predictions are wrong then Im wrong so please can you point to me thoses tests over long periods that I can settle my case. Ill read and report back here.

With a fair ABX test you should be able to listen to and compare samples A, B and X for as long as you want, as many times as you want and in whatever order you want. Only once you have finished listening to all three samples as much as you want and you're completely satisfied do you make the choice: Is sample X the same as sample A or sample B?

Read here: Clicky

Nowhere does it say that there should be a limit on the amount of time or number or repetitions before you make a choice.

When you do an ABX in Foobar you can listen to the samples as much as you want and it even gives you the option to compare specific parts of each track before you make a decision. If you want to compare A to X ten times then B to X ten times or even compare A to B before you make a decision you can. That's how ABX works. That's why it's the fairest way possible to make a comparison.

 

[pyrrhon]

This one just for us : blues the healer !!!!!!!!!!

https://www.youtube.com/watch?v=x0isR1emmPs

Gonna have to enlarge the us circle to you too !

 

[davedotco]

...The problem is simply setting them up and performing sufficiant repititions for the results to be viable, boredom, fatigue and just plain impracticality cause all kinds of problems.

This will only happen if you place a time limit on yourself.

If while doing an ABX you start losing concentration you can stop, go off and get a cup of tea or whatever before having another listen. There is no pressure or time/repitition limits before making a choice.

If you wanted you could even spend just 60 seconds a day over a period of several weeks comparing audio samples before making the choice of which sample is the same as X. Sure this would draw the test out considerably but it would be no less valid.

You could even listen to A every day for a week then B every day the next week then X every day for a week before making a choice. It is still an ABX. You only make the decision once you have listened however much/little you want and in whatever way you want.

True enough, though it does bring you right up against the practicality issue!

If the switching is done by a third party as I think it would need to be in an ABX, then it is hard to see how this could be extended to days. You might be able to set up the comparison using 'black box' switching but you would have to make sure the whole system was tamper proof, that the switch gear was transparent etc, etc, awkward.

 

[steve_1979]

...The problem is simply setting them up and performing sufficiant repititions for the results to be viable, boredom, fatigue and just plain impracticality cause all kinds of problems.

This will only happen if you place a time limit on yourself.

If while doing an ABX you start losing concentration you can stop, go off and get a cup of tea or whatever before having another listen. There is no pressure or time/repitition limits before making a choice.

If you wanted you could even spend just 60 seconds a day over a period of several weeks comparing audio samples before making the choice of which sample is the same as X. Sure this would draw the test out considerably but it would be no less valid.

Or you could listen to sample A all day every day for a week, then B every day the next week then X every day for a week before making a choice. It is still an ABX. You only make the decision once you have listened however much/little you want and in whatever way you want.

 

[steve_1979]

...The problem is simply setting them up and performing sufficiant repititions for the results to be viable, boredom, fatigue and just plain impracticality cause all kinds of problems.

This will only happen if you place a time limit on yourself.

If while doing an ABX you start losing concentration you can stop, go off and get a cup of tea or whatever before having another listen. There is no pressure or time/repitition limits before making a choice.

If you wanted you could even spend just 60 seconds a day over a period of several weeks comparing audio samples before making the choice of which sample is the same as X. Sure this would draw the test out considerably but it would be no less valid.

You could even listen to A every day for a week then B every day the next week then X every day for a week before making a choice. It is still an ABX. You only make the decision once you have listened however much/little you want and in whatever way you want.

True enough, though it does bring you right up against the practicality issue!

If the switching is done by a third party as I think it would need to be in an ABX, then it is hard to see how this could be extended to days. You might be able to set up the comparison using 'black box' switching but you would have to make sure the whole system was tamper proof, that the switch gear was transparent etc, etc, awkward.

Very true. Practicality can be a big issue with ABX tests.

It's simple enough when comparing digital music files because you can use a program such as Foobar to make sure it's a proper and fair double blind test (it won't let you cheat and computers don't have emotions or body language that might subconsciously give it away and skew the result). But when it comes to comparing things like physical HiFi equipment things get more complicated.

Mind you, you can still use computer programs like Foobar for comparing some HiFi equipment by simply using a very high quality, ultra low distortion audibly transparent ADC. When comparing things like CD players, DAC's and cable interconnects you can just record their output into a lossless digital audio file using the ADC. You then use these digital audio files to do an ABX test in Foobar. This way you are effectively doing the ABX test of the CD player/DAC/cable in Foobar. Provided the ADC and lossless audio files are of a high enough quality not to introduce any audible distortion this would be the easiest way to do it with minimal room for any errors or complications.