Quote:
Shiv, if you do a search here on Head-Fi, you'll find a good deal of discussion about that "study." |
MacDEF, I did some searches but had bad luck finding this discussion. You must be talking about
the threads on Headwize.
Here are some of your objections that I found.
1) The test did not define what it meant by "audiophile" -- the listeners could have had vastly different listening levels and skills.
Well, you're right, it was a bit vague about this. The selection wasn't, though, as you claim on Headwize, completely random. C't did first ascertain listener's "qualifications" before including them. They certainly ought to have included those qualifications in their article. However, I think there is plenty of evidence that these listeners were indeed well qualified.
Indeed, the article mentions specifically that included in the test were:
* an sound engineer whose profession it is to prepare classical music masters for production for Deutsche Grammophone
* a student electronics developer well-versed in audio circuitry
* an engineer who worked on the AAC algorithm
* an engineer who develops hearing aids, works on audio signal processing algorithms, and "is used to participating in complex sound tests, mostly dealing with finding artifacts and sound differences"
* an audio lover who owns a 40,000 DM audio system
* a blind listener with perfect pitch who had been involved in the creation of special microphones
Given the high calibre of these participants, there is no reason to believe that the rest are much less qualified.
Even forgetting these others, the poor performance of just these top listeners speaks volumes. The poor performance of even the TOP listener speaks volumes.
2) The test played the CDs and MP3s from a "a PC's cheap sound card."
This simply isn't true. According to the article, "we had the test samples encoded with MusicMatch 4.4 for Windows in joint-stereo, converted into AIFF format with a Power Mac G3 for the Apple QuickTime Player, and then burned onto a single Audio-CD in a random sequence along with the extracted CD Audio files." This means that they played the MP3s on the Marantz CD player.
3) The study method of testing 128k, 256k, and full CD samples simultaneously potentially obscured the differences between 256k and full CD samples.
This is a valid objection, but I think this just goes to show that, whatever else they may be, the differences are not "easily detectable" on a "good system." If one cannot consistently hear these differences on a Orpheus with excellent associated equipment even allowed plenty of time, attention, and effort merely because of the presence of one poorer sample, then the differences are not easily detected. Easily detected to me means that you don't have to strain to hear the difference -- it is heard merely by looking for it.
Additionally, the participants had free reign to switch between 256k and CD samples and compare them exclusively as they wished. Considering that many of the participants would be familiar with listening tests, I suspect this is a strategy some of them adopted: first isolate the 128k, and then duke it out between the two others. It's only logical if the 256k and CD samples were as subtly different as your theory suggests. Obviously we can't know for sure, but I wouldn't so underestimate the intelligence of the participants.
Are there any other problems with the test that you see?
--
None of the objections above explains why some 128k MP3 samples were consistently preferred to the full CD samples of the same music, or why the DG engineer thought that the 256k samples sounded "rounder" and "more pleasing" than the full CD samples.
If indeed a difference can be heard between 256k and full CD samples, perhaps it lies in a completely different direction than expected?
This is more evidence that some MP3s really DO sound better than full CDs on headphones. Does this make *them* the true high-fidelity choice?
--
Finally, recall again that all the discussion above has been about 256K samples. Again, 320k samples logically would have even less of an audible difference from the original CD, if indeed any difference were audible at all.
--
Quote:
Hmmm....another potentially interesting and informative thread gets hijacked and turned into a debate that must have been discussed a hundred times here.... |
Tell me about it--those guys just keep pounding away at that work system argument that'll never end. Don't they ever learn?