Qobuz vs. Tidal vs. CD: Is There a Real Difference in Sound Quality?

#1

Gentlemen,

Without going into details, I would like to ask everyone who can to briefly share their opinions on the following questions:

  1. Do you hear a difference in the sound quality of tracks purchased on Qobuz versus the same tracks streamed directly on Qobuz?

  2. Can you distinguish or hear a difference between the same tracks (released by the same record label) on Tidal and Qobuz?

  3. Can anyone finally answer the question: is there a difference in sound quality between an original CD, Tidal, Qobuz, and a purchased track on Qobuz?

In all cases, of course, we are assuming the same recording label.

In my case, my answer to the first two questions is a definite no. Although, a year ago, I felt there was a noticeable difference between purchased material and the quality of streaming services. Now, I simply can’t hear a difference—the quality seems identical across the board. It’s impossible to tell them apart.

Do you feel the same way? Has it always been like this, or am I imagining things? Or maybe our Hi-Fi ROSE devices reproduce audio so well that they eliminate these differences?

I invite everyone to share their opinions—preferably in a friendly manner!

#2

No.

Possibly, because they could bne using different masters.

Yes, No, Maybe. Streaming services are terrible at telling exactly which master they are using (they are not sitting there, furiously ripping off the shelf CDs) and they change them ever so often.

1 Like
#3

Just one correction - it is important that all tracks are the same master, not only from the same recording label. I am saying this because there are many different masters (usually labeled differently, for expample “master 2022”, or “master 2012”) from the same track. It is also important, when comparing, that they have the same sample rate and bit depth.

That said, I recently have done a null test between Qobuz, Deezer, Tidal, CD rip and Deezer over Chromecast. Deezer in both versions, direct and over Chromecast, were played on Wiim Pro, rest on Rose 250A. Null test showed NO DIFFERENCE AT ALL between all these tracks.

Ok, to be honest, null test omitts timing information, but it shows that all bits received from streaming services on both streamers are bit to bit identical. If we skip discussion about possibility that different stremers can, or can not sound differently if connected to the same DAC, this test, without any need to conduct listening tests, shows that the same master played from different lossless streaming services MUST have identical sound between them and between any of them and a CD rip.

In test was also Spotify, but, as expected, lossy compressed track was not bit to bit identical after unpacking to the rest of the tracks in the test.

3 Likes
#4

The WAV files pressed to CDs use a band-pass filter of 20Hz to 20kHz. If the streamers are working with WAV sources, have those sources already been filtered? I have a CD of the 1812 Overture that goes into great detail about how they captured the cannons down to 6Hz. Obviously, that never makes it to a CD; but do the streaming services use similar filters? How aggressive are they?

#5

How does tht even work, and why do I have CDs with below-20 measurement tracks?

#6

I misspoke. The nominal lower limit on the audio signal is 20Hz, but that isn’t necessarily enforced. A high-pass filter is sometimes applied to get rid of extraneous noise (such as rumble), but the mastering might just rely on the natural roll-off in the recording process. (As far as I know, all of this happens on the analogue side before the A to D conversion.)

A low-pass filter is always needed to make sure the high frequencies don’t cause trouble during the conversion. That gets into mathematics that I won’t try to remember.

So, if the CD is being created for some special purpose, such as testing and measurement, it could indeed include lower frequencies. As for whether or not the playback equipment can do a successful D to A conversion, and then reproduce the reconverted analogue sound, is another matter altogether. That’s the whole point of having a measurement CD.

Apparently, the DAC used by Rose can go all the way down to 0Hz. The CD drive, being digital itself, reads the bitstream off the CD and passes it unchanged to the DAC. If there’s 0Hz recorded on the CD, then theoretically and ideally a 0Hz signal should reach your speakers.

I certainly hope not. 0Hz is a DC voltage, and the last time I saw a DC voltage go through a speaker the result was smoke, not sound.

As I said, my previous post was wrong. It’s entirely possible that all of the bragging in my 1812 Overture’s liner notes is justified. I don’t know how well my subwoofer can reproduce 6Hz. I wouldn’t hear a sound in my ears; I would feel a kick in the chest.

I hope I’ve redeemed myself.

#7

That’s just the commonly used lower range for musical signal that is used for quoting specs.

Yeah, there is a natural upper limit at Nyquist frequency, but there is no lower limit for the frequency that can be encoded.

There might be a DAC that has a trouble at lower frequencies, but it would be something quite esoteric.

Yup. You certainly would not want it, but but any common DAC would happily do it (speakers won’t be as happy, of course) . Quite easy to test with something like Audacity.

And of course there are subs (not even talking about some esotheric fan-type ones that can actually go down to 0Hz) that can reproduce ~10Hz at audible volumes, and there is something in recordings for them to reproduce.

I don’t know which mastering this is, but yes, the problem is that a lot of microphones will not necessarily pick up much at these frequencies) and the rest of the (analog) processing chain is likely rated for 20-20.

Usually, there is very little, if at all, content in musical recording below 20Hz, and most transducers won’t play it anyway, but specifically for CDs (or anything encoded even at Redbook rates) there is nothing saying that you can’t have an arbitrary lower frequency there…

Something like that.

#8

I don’t know which mastering it is, but the liner notes describe how they used special equipment to record the cannons separately. Supposedly their mikes went low enough to cover 6Hz.

Oh, and when I said the playback might be a problem, I was referring to the analogue stuff at the amp and speaker end.

#9

Oh, that for sure, normal mics used for normal music don’t need to go that low. A well-designed amp should not have much problem dealing with 6Hz, but getting any meaningful output from the speakers would be a problem. There are subwoofers that could go that low, but we are talking about something VERY big and VERY VERY expensive (as somnewhat reference, $25,000 JL Audio Gotham v2 weighs 360 pounds and is -10dB at 14Hz).

With headphones you would have better luck actually hearing it, and those can be had for some reasonable money (Focal Celeste, at $599 goes to 5Hz at -3dB). So I guess it was not a complete waste to go to all these length to record that cannon, but for the vast majority of the listeners it is really quite irrelevant anyway.