Separate names with a comma.
Discussion in 'Sound Science' started by bigshot, Dec 8, 2014.
Almost years after the announcement of their launch and they're still stalling? Amazing.
At the risk of providing more treacle, that depends on what you mean by cumulative but I don't agree that improving part of the chain is probably a good thing, it's just as likely to be a bad thing. Let's really simplify it and just use arbitrary numbers. Let's say perfect is zero and the anti-alias filter in a particular ADC equals +1, the reconstruction filter in a particular DAC equals +2, MQA knows that ADC and DAC and therefore applies -3 and we're back at zero/perfect. Massively simplified, this is essentially what MQA are claiming to do and on the face of it, it sounds like a great idea but it's missing what actually happens in practice. Typically numerous anti-alias filters have been applied in any given master. Not only in the recording ADC but in many of the audio processors used in mixing and mastering, some of which over-sample process and down-sample again or occasionally do the reverse. There may be dozens of anti-alias filters applied in a given mix and the interaction is not cumulative the way you are suggesting, using my simplified analogy, some/many of the filters may effectively be producing a negative cumulative effect. A particular oversampling compressor might cause say a -3 effect and be used say 5 times, so the total figure above of +3 could actually be -12 and then along comes MQA, corrects that by -3 and now the error is -15, even further from the perfect zero, it's made matter worse, not better. The effect of multiple anti-alias filters and other processors which affect timing being applied at the different stages of recording, mixing and mastering are unpredictable and therefore applying the correction stated by MQA is just as likely to be worse than better. That's not so much of a problem though because "on the face of it" is all that's required, audiophiles make a lot of fuss over reconstruction filters while apparently being completely ignorant of how insignificant that is compared to everything else which actually happens in the production of the music.
The listening test is in. The good news is that MQA didn't degrade the sound quality at all. The bad news is that MQA didn't sound any different than the PCM.
The first tranche of Archimago's test results are now available.
BREAD AND BUTTER! JINX YOU OWE ME A COKE!
Well I have to say I am not surprised.
I welcome any advance in audio, but this does look like it is the significant jump we were promised.
I'll leave it to others to bang on that this is only MQA core and the full version is better.
The final subgroup breakdown has been posted. It seems that it's pretty much a draw again. Looks to me like a random spread.
I don’t see a statistically valid conclusion one way or another, the results appear more or less random. I thought it was kind of interesting that young people and people with high end systems, two groups I imagine would be able to hear the details best, both preferred PCM. But not to any statistically meaningful degree.
The lack of MQA software encoders on the market right now makes me hesitant. If everybody is relying on MQA for test files it's not really a controlled test.
This test is only trying to show that the part of MQA he tried to copy doesn't offer much of an audible impact. Which if not conclusive or properly controlled, was at least expected.
Now would other manipulations in a typical MQA album, result in audible differences? And if yes, are they really consistently preferred like marketing wants us to believe? This test doesn't say. And probably no test will as MQA people are right now the only ones who could help to set up a proper test, and they have no interest in actually offering a fair opportunity for comparison. Because we know there would be none or at best critically small ones that might not be an improvement. MQA will benefit much more from people going crazy over what are simply different masters. You know, the "happy" misunderstanding encouraged by all formats.
That hasn't done much for SACDs and blu-ray audio.