A Comparison of Clarity in MQA Encoded Files vs. Their Unprocessed State as Performed by Three Groups — Expert Listeners

Let me quote from article ‘The researchers were interested in comparing the responses of the three target groups and whether playback systems had any significant effect on listeners’ perception. Data shows that listeners were not able to significantly discriminate between MQA encoded files and the unprocessed original due to several interaction effects’.

http://www.aes.org/e-lib/browse.cfm?elib=19396

Finally, results from a proper scientific study… At least as far as clarity is concerned, AES’ listening test results show that there’s no statistically significant difference with MQA vs. source PCM. So, all the BS from MQA, Stereophile, and TAS about how amazing it sounds etc, is bogus or just in comparison to low-res MP3?

Anyone is a AES member? Can download for free.

2 Likes

Shame they can’t hear what I hear :joy:

Data shows that listeners were not able to significantly discriminate between MQA encoded files and the unprocessed original due to several interaction effects.

What am I missing here? Isn’t the above finding in favor of MQA? Or does MQA claim to sound better than original hires? (excuse my ignorance, don’t know much about the thing despite having tried to follow very lengthy threads here).

“due to several interaction effects”

Ordinarily it would seem sufficient just to conclude that the deviation of listeners identification from random (the null hypotheses) was not statistically significant. The above line sounds like a qualification indicating a problem with the methodology.

Not necessarily. I’d have to see the paper, but in a typical ANOVA statistical analysis there could be main effects and interaction effects that are legitimate and don’t indicate a method problem.

Thanks a lot. Now, I’ll have nightmares about not being ready for my experimental psych final.:laughing:

3 Likes

I respect your opinion!
But analysis like that is just a waste of time!
Music is for listening to and have a lots of enjoy :blush:
Put all time on testing and analysis on cancer and other horrible kinds of sickness instead, so a cure can be found!
That’s my opinion!

Love & Respect

2 Likes

@Anders_Strengberg. I’m not pushing the idea that we need to be doing research on MQA (or audio listening in general) in this context.* My comment was only to make the point that in research designs in general (including research related to Cancer), that it is often appropriate to expect an interaction effect as well as main effect.

So we don’t necessarily disagree. I was just pushing back on the statement that for research in general an interaction effect meant a flawed design or bad results.

*Edit: I don’t mean that professionals shouldn’t be doing research to improve audio hardware and software. Heck, we wouldn’t have ROON without that. I mean that as end users (listeners), I’m happy to enjoy the music. And the format of the music has little or no effect for me, beyond some base level of quality.

1 Like