CD vs SACD Sampling Rates
Tom Martin -- Sun, 05/04/2008 - 08:26
Mix Magazine recently published an interesting article on whether people can actually hear the difference between CD and SACD:
http://mixonline.com/recording/mixing/audio_emperors_new_sampling/index....
Of course this used blind A/B testing. As TAS' Robert Harley has written in an Audio Engineering Society paper, there are many reasons to doubt the usefulness of this technique for subtle phenomena.
This "test" says more about the flaws of blind listening tests and the agenda of the "experimenters" (both rabid "objectivists") than it does about the audibility of high-resolution digital audio. Over the years, blind listening tests have suggested absurd conclusions. For example, one test "revealed" that an output-transformerless tubed amplifier, a massive high-end solid-state amplifier, and a $220 receiver all sound the same.
Robert Harley
Maybe they did all sound the same?
Have you listened to CD, high-res PCM, and DSD and concluded that they all sound the same?
I thought this subject was pretty much exhausted. Double blind tests are structurally flawed and there is nothing anyone can do to change that fact. The ear brain system is highly sensitive to context and state; I have yet to see any thoughtful response to this structural flaw. Context refers to other information provided to the brain taht might influence hearing like price, brand reputatuion, the physical appearance of the component, on and on. Multiply taht with the "stress" induced in such tests. "State" refers to things like hours of sleep, alcohol, caffeine, mood, etc.
Thank god for "observationists" whose confidence in their ears have helped uncover breakthrough discoveries like jitter, oversampling, and the impact of HF extension on dynamics in the audible range; all of these would have been (and were) rejected by "objectivists" till we found ways to measure these phenomena. Now of course it is "objective", since we found a way to measure them. How many more discoveries of important new things to measure lurk in the "yet to be known" zone, I wonder.
In a scientifically valid double-blind comparison of sound A and sound B, there are only two unbreakable rules: (1) the volume levels must be matched within 0.1 dB and (2) nobody is allowed to know in advance which is A and which is B. Those who deny the validity of double-blind testing need to answer the following question: WHAT SPECIAL AUDIO INSIGHTS ARE TO BE GAINED BY NOT MATCHING THE LEVELS AND PEEKING AT THE NAMEPLATES? Please don't give evasive answers; list those special insights or admit that there aren't any.
Peter:
Do you still think that after-market power cords are a "waste" ? Have you compared (newer) high-end designs to mass-market ones? Everyone (except you) says they're dramatically improving the sound of their electronics these days......
How about signal paths, Pete ? Do they all still "sound the same", if not "pushed" ? They might...with string quartets played-back at lower output. But one of the main goals of a high-end system is to play large-dynamic-swing music back at realistic levels. There can't be anyone in the world (besides Floyd Toole) who believes this "same sound" BS - and even *he* might have had a change of heart by now.......
.
Post new comment