As I mentioned in my post to Tony below, I've been using the software tool, Spek, to examine the frequency spectrum plots of some 24/96 files I've recently downloaded. Since I'm a newbie at this, I'm going to attempt to interpret what I'm seeing, and if I'm wrong, please feel free to disabuse me of my mistaken notions!
The first plot is of the final track from the Acousence recording of Stravinsky's "Rite of Spring" with Jonathan Darlington and the Duisberg Philharmonic, which I downloaded from the Linn site. This is a really great and powerful recording of this work - never mind that you've never heard of this orchestra before! This company records ONLY in PCM, so I don't believe there has been any DSD/PCM conversion.
As you can see, the frequencies get up to about 30KHz, even though the 96K sampling rate allows for a theoretical maximum of 48KHz, as can be seen on the scale on the left hand side of the chart. There are of course many factors which could be limiting the extension.
One limiting factor is often the dynamic level of the music itself. If the music is quiet, one tends not to experience the higher frequencies, since the higher overtones are so much quieter than the fundamentals, which are already quiet. To illustrate this point, here is a plot of the Acousence recording of the second of Turina's "Danzas Fantasticas", a relatively placid composition, with the Anhaltische Philharmonie, Dessau, also downloaded from the Linn site:
So even though this is a 24/96 recording, we don't see much above 20KHz, aside from the opening flourish.
So far, everything looks pretty much as expected. However, here is an interesting excerpt from The Classical Shop site of the Chandos recording of Weinberg's Symphony No. 3, again at 24/96 - this BTW is the orchestral excerpt that users can download for free to make sure the download process works as it should for you on this particular site. This is interesting music, performed by the BBC Philharmonic:
What do we see here? Nothing over 25KHz, except way at the top of the scale, where we see a dusting of supersonic frequencies, disconnected from the music signal at the bottom. In contrast to the other plots, the master for this recording was DSD (as noted on SA-CD.net), and this plot allows us to see the beginnings of the supersonic noise characteristic of that recording method, just as Tony mentioned in his post below.
So I guess my final question is, am I interpreting what I'm seeing in these plots correctly? Am I missing anything of interest?
This post is made possible by the generous
support of people like you and our sponsors:
