I assume MOST of these tracks are 44/16 redbook tracks that the streaming service is upsampling, right?
Sure, newer tracks might be mastered in 192 or 96, but the vast majority are redbook tracks, right? (I mean, lots of albums are available in both, and I don’t think they’ve been remastered for 192, and are just being upsampled.)
Just curious if Roon’s upsampling does the exact same thing.
Lately I’ve been upsampling to 256 DSD, and it sounds great. Just trying to justify the hi-res tier from the streaming services.
I am using both, depending on zone or system/streamer.
I use HQP for my raspberry pi zones… They have Pi2AES HATs, and those only go to to PCM 192.
But my main system has a (fairly recently added) eversolo DMP-A6 streamer into my Denafrips Pontus II, and that can handle DSD256. I use roon to upsample for that zone.
Gotcha. Wasn’t sure. There’s another thread on this (which you also replied to!), which states the streaming services get the hi-res (remastered) files from the studio.
Doesn’t that have a USB C or even 2 USB input’s or am I confusing it with the DAC
Either way the difference is not night and day, but I do prefer HQPlayer upscaling to Roon.
Though I am mostly happy with 16/44 for most of my listening.
But I have recently purchased a number of 24/192 items in the Qobuz sale, not because I needed the quality but because I could and they were cheaper than 16/44.
I do wonder how these are remastered and transferred from 50/60 year old tape before being converted to digital. But when they sound so good I am just happy to accept it
They most probably capture the tapes in digital before re-mastering. If they capture master tapes, I would think the only ‘mastering’ would be to scale and down-convert the digital capture to various resolutions. If they capture raw tapes, then they can do proper digital mastering.
I’ve not looked at spectrograms for hi-res tracks from streaming services, but I have for digital download services. The result was a mixed bag: some tracks looked upsampled, evidenced by a sharp drop off in energy above 22.05kHz; others had a smoother decrease in energy, suggesting that A2D had been done at a high sample rate originally.
I’d say that selling “hi-res” tracks that are really Redbook upsampled is deceptive. But for the most part, I’ve only seen this for old material where the chain of custody from original tape to digital download might be iffy. I’ve never seen a recently recorded track with this issue.
You never know. Some files are true HiRes, some of them are only upscaled. On the AES a paper was presented in 2015 how you can detect it. Why does the AES present something like that?
This is what the site prostudiomasters.com states: Before I got qobuz, I made several purchases from this site. Now with Qobuz sublime, hard to justify. Vey much qobuz like site with a wide range of music, but focus on jazz and classical. These tracks are high resolution tracks are made from the original analog master. These are very high quality labels. I am sure there is some upsampling, but not all. Some labels remastered their tracks into high resolution.
℗ 2023 The Nina Simone Charitable Trust and Rich & Famous Records, Ltd, under exclusive license to Verve Label Group and UMe, divisions of UMG Recordings, Inc; A Verve Records / UMe release You’ve Got To Learn (Live)
192 kHz / 24-bit, 96 kHz / 24-bit PCM – CM BLUE NOTE (A92) Studio Masters
Recording by Rudy Van Gelder. Photo by Francis Wolf. Cover Design by Reid Miles. High-Resolution Mastering by Alan Yoshida and Robin Lynn at Blanche Dubois, April 2012.
When the Universal archives burned, many master tapes were only saved because they were not in the archives at the time, being shipped around the world for hi-res digitization / remastering jobs, or the masters did burn but had already been digitized. So it seems that indeed much of this is legitimately going on.
There’s definitely data up to 25 or 30Khz. Whether that’s original data from the master tapes or the result of upsampling, there’s no way to tell. Here’s a screenshot from the CD for comparison.
One of the things MQA attempted but never got right is provenance. It would be great to have this information, but I don’t think anyone in the music industry or the general public cares. It’s only us OCD audiophiles that care. The only reason we have HD music is because it’s something new the record companies can sell. If there was a legitimate market for music with documented provenance, the record companies would sell it.
initially i thought the screenshots were reversed, as it looked like the CD had a higher dynamic range… but I zoomed in on the Y axis and I see the difference.
whether we can hear that difference, I guess is the question…
Hum, that steepish drop off at around 22kHz uniformly across the track makes me wonder… For comparison, I’m showing below the analysis by (sadly no longer supported) MusicScope on a recent multi-instrument jazz track, 96/24. Notice the smooth energy drop, and the lack of a uniform “cliff” at 22.05kHz.
FLAC 192 kHz 24-bit*
