The only SACD I own that I can't tell any difference between the CD and the SACD layer is this one.... Air Supply - The Definitive Colleciton
It looks like that is a SACD released only in Hong Kong, if you can trust SACD.net's information. It would not surprise me if the DSD layer is nothing more than upconverted from redbook PCM. Hong Kong labels have had a habit of releasing CD masterings as SACDs on occasion.
With all the discussion herein about theoretical and realized differences among different formats, it's nice to have some real-world data. Thanks. On an experimental level, there are certainly quantifiable differences in sound among CD, SACD, HDCD, and so on, but to the average listener they probably really don't matter a whole lot. Moreover, I suspect mastering approach matters more, followed by individual bias/preference. Some actual double-blind studies would be helpful in this regard.
Optimal age for hearing of a human being is approx 5-19. Worst ages for human hearing = 40 then exponentially worse each couple of years until the end. Approx median age of those looking to spend money on advantages of SACD = 58. % of those who lay out what they consider good money for quality SACD gear that will truly believe they do hear huge differences = 99.95%. Amount of times for those over age 40 that in actuality that will be true = 0 Folks on this forum that will argue all the above to their bitter end = 80% % of statistics that are seemingly pulled out of ones posterior = 75%
Intriguing study that I will have to check out. I think you are correct that a mastering engineer for MFSL, for example, would feel obliged--or pressed--to put in a top-notch effort. This could thereby skew the general opinion toward SACD being a superior format when we are simply hearing a better mastering approach within that format. This doesn't mean one shouldn't spend the extra dollars on SACDs because they may still offer the best available sound of a particular set of music.
While I do agree, to some and on some gear DSD has sonic signature that some like or dislike. That aside any who are 40 or over receive no benefit as only the very very lucky will still be able to hear even the tiniest amount above 17k - knock that down to the very very lucky at 15k at age 50 and 13K at age 60 In terms of the engineer being able to shape frequencies in human hearing range better if done in hi-res domain - OK fine and true, but the listener is still only hearing at or below the freq they are still capable which are 15k/16k/17k/18k/19k/20k/21k/22k depending on age. While the pure technical merits of SACD are vast and empirically provable, the facts of human hearing limitations are 99.9% of the listener equation . Every human is affected by "hearing loss consistent with age", anyone who thinks they are not is wrong or had better be part Canine or Kryptonian.
I don't know that looking at someone's ability to perceive higher frequencies fully answers the question. If my understanding is correct, SACD has better signal to noise ratio at lower frequencies than CD, but actually is worse as the frequency increases. If the SNR is better at midrange frequencies and below, where most listeners still have the ability to hear the sound as presented by the speakers, and the noise is worse above 10kHz where a lot of people cannot hear well anyway, that could explain SACD sounding better to some people but not others. But I'm a big believer of deciding for yourself by listening whether any difference is "worth it."