CD vs HD download question *

In studio production, editing, mixing, mastering choices are related to being able to edit, change levels, noise levels, and many other things, that have no relation to simple in home playback of audio.

 

I would like many better recordings and masterings!

 

Yup. I'm going to do some comparisons of SACD vs. Redbook layers on the same disc.

for consumers, what really matters is to look for well-regarded releases, SACD, HD Tracks, CD, or whatever and get the best masterings/recordings

 

On well done Rock recordings I can easily spot the Hi Res version just by listenning to drums, 44.1/16 does something to drums that gets them sound unnatural both pace and timbre wise. In my opinion, only talking about well done recordings, drums sound more natural with good pace/timming and fine timbre on vinyl and High Res digital.
Lets never forget than anti-aliasing filters get IN the audio band on 44.1 or 48 sampled recordings causing phase shifting and high frequency attenuation. With high sampling frequencies like 96 or 192 Khz the undesired effects of the anti-aliasing filter are far from the audio range. Just my 0.02$.

 

Note that hybrid SACD's RBCD layer tracks are frequently mastered inferior to DSD layer ones, i.e.:

Roxy Music - Avalon (Hybrid SACD)
Scorpions, Berliner Philharmoniker - Moment Of Glory (Hybrid SACD)
Rachel Podger, Arte Dei Suonatori - Vivaldi: 12 Violin Concertos, Op. 4 'La Stravaganza' (Hybrid SACD two-fer)

& etc.

 

Yeah, there's no guarantee the CD layer comes from the same master just downsampled on a SACD. They seem to often come from completely different masters...

 

Not sure what the Naxos CD mastering is, but 16/44 FLAC files provided together with eClassical's 24/88 FLAC versions are seemingly a straight forward linear-phase resamples to 44kHz & downconersions to 16-bit using some kind of extreme SHAped dither from HD masters (see null-test waveform spectra below):


I made my own CD version of the album using MDA dither VST plug-in for fb2k & there is no way for me to tell the difference between my downconversion & the HD master (see track spectrum deltas below; S-TPDF dither is included for comparison):

 

I've done several DSD to 44.1/16 downconversions (from my ripped SACD's) using Weiss Saracon adding TPDF dither and even 'though Saracon is a damn good converter, the differences are there. I did this with Telarc's The Sound Music Of Jerry Goldsmith (a native DSD recording) and Analogue Production's The Planets conducted by Zubin Mehta (originally released on Decca Records in 1972) and believe me, with minimal trained ears one can tell one from the other. But, with the same ears DSD to 176.4/24 conversion of both albums using Saracon are almost undistinguishable. I'm agnostic by nature in all senses thus I believe in FACTS rather than theory.

 

Not sure if you saw this study posted earlier in the thread, but it was quite eye opening to me.
I had not seen it before and had wrongly assumed different results were more common.
AES E-Library » A Meta-Analysis of High Resolution Audio Perceptual Evaluation

FREE to downlaod, but these studies even with trained listeners were only able to reach approximately a 60% or slightly better ability to identify High res Samples.

 

only 60%? like that's bad?

Here's the summary of the study from that link:

 

Have a shot at the test I mentioned in another thread & post your successful (i.e. 9+ correct answers out of 10 attempts) ABX test report...

 

One important consideration - D/A converters don't behave the same at all sample rates and bit depths, in part due to their analog components. Also, many older D/A converts have subtle bugs with high res audio, which can cause them to sound different when playing back high res audio. So, one D/A converter might sound quite different with a high res version of an album, whereas the other could sound virtually identical at 16/44.1.

 

"Results showed a small but statistically significant ability of test subjects to discriminate high resolution content, and this effect increased dramatically when test subjects received extensive training. This result was verified by a sensitivity analysis exploring different choices for the chosen studies and different analysis approaches. Potential biases in studies, effect of test methodology, experimental design, and choice of stimuli were also investigated. The overall conclusion is that the perceived fidelity of an audio recording and playback chain can be affected by operating beyond conventional levels."

Well 50/50% is random guessing, so 6 times right out of 10 is better than guessing, but for sure nothing definitive ..... and that was with the trained listeners.
I was hoping it would be far higher.

 

I may try a graphical comparison of the actual CD to the high-res, since I have the CD. What software can do it?

 

I got 4 hours ago as a Christmas present the CD for the Star Wars The Last Jedi soundtrack (I'm tired of telling my family "don't buy me a CD as a present unless I tell you so, I don't like them") and I bought a couple of days ago the 192/24 downolad of the same album. Maybe they are different masterings or maybe they are sourced from the same mastering and the CD got one final step which would be converting 192/24 to 44.1/16, the fact is that the Hi Res download sounds much much better than the CD, soundstage is wider, separation between instruments is more noticiable, timbre sounds more realistic to my ears, there's "air" on the sound while the CD sounds artificial, "congested" and if there was a sheet between the speakers and me.

 

The CD mastering is slightly different & uses a tad of compression (or peak limiting), noticeable on HD master tracks with peaks approaching 0dBFS (RMS stays the same for both HD & CD tracks, while peak levels on several CD tracks drop by as much as 0.44dB compared to their HD counterparts). Who knows what resampling + dithering algorithms were used to produce CDDA master also...

 

Here is my older comment about the tools I use