• Welcome to the all-new HUG. All content has been converted from the old system, and over the next days we will re-style HUG in a more familiar way.

Streaming must be better than CD?

Just purchased Spotify premium and I would like to know other experiences with streaming.

From what I have heard yet and compared with my own cd's I have the idea the quality of streaming is at least equal or better.

It could make sense because the quality of the cd player still depends on mechanical parts,the laser and data is not (or less) converted ( true ?) versus microprocessing/streaming.

Is streaming closer to the source ?
 

Kumar Kane

New member
Only as good as the source

Only as good as the source

Is streaming closer to the source ?
Close enough in most cases for it to not matter, if you are able to obtain streams that are 256kbps or higher. Even where that isn't the case, streams of 128kbps and higher are good enough, unless instead of enjoying the music, you are listening for what is wrong with the quality. But I would not say it is closer than a decent CDP that is available these days for less than USD 100. These now perform very well indeed.

And a poorly mastered recording will sound poor via both your CD player and via Spotify. Unfortunately there is a huge amount of such music out there, but Spotify isn't the culprit or the solution provider for that state of affairs.
 

Nessuno

Member
Lossy compression - inaudible?

Lossy compression - inaudible?

Is streaming closer to the source ?
Short answer: no.

Actually, it depends on if the streaming service is applying any lossy compression or not. Spotify for example, uses ogg-vorbis lossy compression.

Anyway the very fact that, not knowing this and so having no expectation bias, you found streaming subjectivley on par with CD proves that the lossy compression used is transparent to your ears, so maybe you better forget this issue and just enjoy music! :)
 

Jeff_C

Member
What quality is Spotify Premium?

What quality is Spotify Premium?

Just purchased Spotify premium and I would like to know other experiences with streaming.

From what I have heard yet and compared with my own cd's I have the idea the quality of streaming is at least equal or better.

It could make sense because the quality of the cd player still depends on mechanical parts,the laser and data is not (or less) converted ( true ?) versus microprocessing/streaming.

Is streaming closer to the source ?
If Spotify Premium is streamed at 16/44.1 the same as CD standard ( I do not know if it is but I suspect it is less than CD quality) then the results should be much the same as when a CD player outputs its audio digitally to the same DAC that is converting the Spotify stream. There are probable technical variables such as the amount of jitter but I doubt that would translate to an audible difference in a listening comparison.
 

Nessuno

Member
Jitter, but not source related

Jitter, but not source related

If Spotify Premium is streamed at 16/44.1 the same as CD standard ( I do not know if it is but I suspect it is less than CD quality) then the results should be much the same as when a CD player outputs its audio digitally to the same DAC that is converting the Spotify stream. There are probable technical variables such as the amount of jitter but I doubt that would translate to an audible difference in a listening comparison.
Upon using the same DAC for both sources, jitter may differ only if the last transport medium used to feed digital data to the DAC is different, for example S/PDIF optical versus USB.

But this of course doesn't depend on the source at all.
 

Nessuno

Member
Ripped from CDs?

Ripped from CDs?

By the way, about streaming compared to CD, quality wise: a lot of Spotify tracks (I can only speak of them as this is the only streaming service I use) are evidently ripped from CDs and a few of them very very poorly ripped!

This is the only real disappointment I have with the sound quality of their service.
 
Spotify often poor quality

Spotify often poor quality

By the way, about streaming compared to CD, quality wise: a lot of Spotify tracks (I can only speak of them as this is the only streaming service I use) are evidently ripped from CDs and a few of them very very poorly ripped!

This is the only real disappointment I have with the sound quality of their service.
You are right, the quality is sometimes poor and this happens too often for me to invest in this application, maybe they focus more on popular music.
 

G Spiggott

New member
Prefer Spotify

Prefer Spotify

I am more than happy with Spotify Premium, but as it is lossily encoded it cannot be as good as CD - all other things being equal.

It is not true that CD is in any way dependent on the quality of the mechanical parts, as long as they reach the minimum standard. They can be made of plastic and flimsy bits of metal for £10 and the result is indistinguishable from the £10,000 'CD 'transport' that audiophiles drool over. If this is not true, then digital audio doesn't actually work - but it does.

An integrated CD player is the same as asynchronous USB or the playing of audio files from memory within a PC: conceptually the audio data isn't 'pushed' to the DAC, but 'pulled' into it. The DAC requests the data in chunks and this is placed in a FIFO buffer (First In First Out) and read out at a precise rate at the DAC's fixed sample rate.

S/PDIF and isochronous USB are different: here the DAC has to adjust itself to the incoming data that is, indeed, being 'pushed' into it at a rate defined by the 'pusher'. Clearly this method is more practical in some situations, but is not quite as conceptually perfect as the one where the DAC's sample rate is fixed and it 'pulls' the data in as needed. CD (in the form of an integrated player) is such a system.
 

Kumar Kane

New member
Excessive quality, a waste

Excessive quality, a waste

I am more than happy with Spotify Premium, but as it is lossily encoded it cannot be as good as CD
From an engineer's point of view, it cannot of course be the same. But what is good? In audio, only if it can be reliably identified as such in a blind comparison test done in a manner that only allows the working of your ear to be an input. By this test, lossy can often be just as good as CD. Returning better performance that can only be seen on an instrument is irrelevant to how good CD is. Quality is determined by how well the purpose something is supposed to fulfil does that, and more quality than that is a waste of resource.

Just like CD is just as good as supposedly better hi res formats. But which are a wasted engineering resource.
 

G Spiggott

New member
From an engineer's point of view, it cannot of course be the same. But what is good? In audio, only if it can be reliably identified as such in a blind comparison test done in a manner that only allows the working of your ear to be an input. By this test, lossy can often be just as good as CD. Returning better performance that can only be seen on an instrument is irrelevant to how good CD is. Quality is determined by how well the purpose something is supposed to fulfil does that, and more quality than that is a waste of resource.

Just like CD is just as good as supposedly better hi res formats. But which are a wasted engineering resource.

I would happily challenge anyone to spot a problem with Spotify Premium - I am confident they wouldn't. But I think there will always be signals that encoders with high-ish compression ratios have difficulty with. No listening test can guarantee to include all those 'problem signals'. Unlike the analogue signal degradations we are used to, lossy encoder problems will be very sudden and 'nonlinear'.

For Ogg Vorbis (the encoder used in Spotify) there can be artefacts:

...the most consistently cited problem with Vorbis is pre-echo, a faint copy of a sharp attack that occurs just before the actual sound (this artifact is most obvious when reproducing the sound of castanets).

When the bitrate is too low to encode the audio without perceptible loss, Vorbis exhibits an analog noise-like failure mode, which can be described as reverberations in a room or amphitheater. Vorbis's behavior is due to the noise floor approach to encoding; see technical details.
https://en.wikipedia.org/wiki/Vorbis

I don't listen for artefacts on Spotify, though. I have heard a strange and persistent artefact on one track, but the track was available on two different Spotify albums and was OK on one of them, so I suspect the other was derived from a lossily-encoded source already and perhaps we were hearing the effect of double encoding or some such.
 

Kumar Kane

New member
Lossy as good as source

Lossy as good as source

the track was available on two different Spotify albums and was OK on one of them, so I suspect the other was derived from a lossily-encoded source already and perhaps we were hearing the effect of double encoding or some such.
It does look like there was some issue with just one implementation if the other was ok.

On the Vorbis comment - what was the threshold for the low, as in how low does the bitrate have to be for the stated failure to be exhibited? I could not find that anywhere in your link.

I have purchased loads of extremely well recorded music via iTunes in the last couple of years, and all of it in 256kbps AAC lossy. I have never once felt: I wish I had that on CD. And this, on some very decent 2 channel set ups. Perhaps my ears are just too old, but I have also heard of very little success in blind picking 320kbps files compared to lossless where both are from the same source material, in a level matched blind test.
 

G Spiggott

New member
Streaming and the audiophile

Streaming and the audiophile

I have purchased loads of extremely well recorded music via iTunes in the last couple of years, and all of it in 256kbps AAC lossy. I have never once felt: I wish I had that on CD. And this, on some very decent 2 channel set ups. Perhaps my ears are just too old, but I have also heard of very little success in blind picking 320kbps files compared to lossless where both are from the same source material, in a level matched blind test.
I am the same. There is an assumption among audiophiles that streaming is automatically 'mid fi' but I do listen intently to music on Spotify, played quite loud. It isn't just a case of a lack of audible 'artefacts': it sounds absolutely fresh and pristine as well.

Some of the music I have on CD also, and I certainly don't detect any difference.
 

Nessuno

Member
Casual listening and compression artefacts

Casual listening and compression artefacts

By the way, about streaming compared to CD, quality wise: a lot of Spotify tracks (I can only speak of them as this is the only streaming service I use) are evidently ripped from CDs and a few of them very very poorly ripped!

This is the only real disappointment I have with the sound quality of their service.
Just to clarify: I was referring to sound quality problems deriving from bad CD ripping: glitches, jumps, tracks abruptly ending, the kind one could experience playing a damaged physical CD in a player (or a good CD in a damaged player).

I have no issues at all with Vorbis compression artifacts. I find it satisfactory starting from what Spotify marks as "high quality" (Vorbis -q5, about 160 kbps). At "normal quality", (Vorbis -q2, about 96 kpbs) I can consistently spot some annoying artifacts, like warbling in harpsichord solo notes. Anyway even the normal setting could be OK for me for casual listening.
 

Nessuno

Member
Ripping problems?

Ripping problems?

Just to clarify: I was referring to sound quality problems deriving from bad CD ripping: glitches, jumps, tracks abruptly ending, the kind one could experience playing a damaged physical CD in a player (or a good CD in a damaged player)..
For example this track at 0:43, 0:53, 0:56, 1:50 etc...

Vor deinen Thron tret ich hiermit, BWV 668
Johann Sebastian Bach • BACH, J.S.: Cantatas, Vol. 9 (Gardiner):

https://open.spotify.com/track/1Px0OqarzWB9JBceAdCB9w
 
Top