CD vs Streamed, by Paul McGowen of (PS Audio)

[georgehifi]

 

Streaming's compressed, no matter who/what you use.

 

Paul McGowan's on the money for a change here. 

First half is right on, second half is an ad for what he has that can part fix it, coming out this month.

But even he rightly says, it's not a complete fix.

 

https://youtu.be/zeHxHaqtKOg

 

He puts the compressing problems  on the streaming companies, I'll go down even one more layer again, and put it on the suppliers of the music to the streaming companies that are also doing the compressing to them, you can see it in my signature "link" below "Yellow Brick Road's" version history of getting compressed through the years all taken from the same original Master

 

Cheers George

 

Edited January 6, 2023 by georgehifi

[Cloth Ears]

When talking about it being streamed from an external source, would have to agree. Although 'compression' might not be the right word in all cases. Even if the source being streamed is an uncompressed digital version of the original master, you have such issues as poor buffering, line-speed, etc. will contribute to less than adequate playback. It's why I prefer to borrow BluRay movies from the library than try to stream them in 4K (and if I like them, I buy the 4K).

 

If the method is complete download prior to playback, then the only contributor to 'compression' (or lack of sonic quality) is the makeup of the file.

Edited January 5, 2023 by Cloth Ears

[bob_m_54]

He's talking about file compression though, not audio (dynamic range) compression. And yes, lossy formats of file compression do compromise the audio quality of the music to varying degrees. Even worse if the source material has had dynamic range compression applied to it before data compression, like you say.

 

Of course if you are streaming lossless compressed files (flac, alac) from your own file server, they will only suffer from the latter type, if your source material has had excessive dynamic compression applied, before you ripped it.

[Cloth Ears]

1 minute ago, bob_m_54 said:

He's talking about file compression though, not audio (dynamic range) compression. And yes, lossy formats of file compression do compromise the audio quality of the music to varying degrees. Even worse if the source material has had dynamic range compression applied to it before data compression, like you say.

 

Of course if you are streaming lossless compressed files (flac, alac) from your own file server, they will only suffer from the latter type, if your source material has had excessive dynamic compression applied, before you ripped it.

I reckon there should be a separate term for when you are listening from your own file-server. 'Streaming' as a concept applies to both - but very different results can occur.

[bob_m_54]

Like maybe "Network Streaming" and "Internet Streaming"?

 

Edit: The same could be said for the term "compression" too "data compression" and 'dynamic range compression"

Edited January 5, 2023 by bob_m_54

[georgehifi]

50 minutes ago, bob_m_54 said:

He's talking about file compression though, not audio (dynamic range) compression.

 

Sorry, Edgar says in the 40th second of the video it's "far less dynamic" and McGowan agrees with him in that.

https://youtu.be/zeHxHaqtKOg?t=40

I say it's more also because of the later compressed version of the masters used as the below Elton John link shows.

Cheers George

Edited January 6, 2023 by georgehifi

[mjs]

Just let me get my can opener and a fishing line for the worms 

 

2 hours ago, georgehifi said:

Streaming's compressed, no matter who/what you use.

 

Well, a FLAC file is a compressed file, doesn't mean you don't get proper lossless playback 

 

It's a good discussion, but a lot of subjective unsubstantiated opinion. Mine is that streaming can give you just as good results as playback of ripped files or CDs. Of course, there is the whole argument about whether we are listening to exactly the same files in each case, streamed or ripped on HDD/CD

[was_a]

Buffering issues? The make-up of the file? Compression?

 

The single biggest factor is NOISE! Streaming solutions are inherently complicated compared to a CD transport. They also negate the sound quality differences of CD pressings—an optimal pressing contributes greatly to audio quality, in my opinion.

[georgehifi]

46 minutes ago, was_a said:

The single biggest factor is NOISE! Streaming solutions are inherently complicated compared to a CD transport.

 

Not just noises from the streaming processes and getting to you from half way around the world, but also from the equipment that the end user plays it back on.

The better CD players have very good multi regulated linear supplies, where as most I know of streamers, have noisy smps in them.

 

Cheers George

Edited January 6, 2023 by georgehifi

[Artnet]

32 minutes ago, georgehifi said:

The better CD players have very good multi regulated linear supplies, were as most I know of streamers have noisy smps in them.

 

There is some severe traffic control occurring on one of your other threads George in regards power supplies   Given I mostly listen to CD I cant complain.

 

3 hours ago, georgehifi said:

Paul McGowan's on the money for a change here. 

 

Agree with you here and have to listen to what he says

 

"bit for bit it looks the same, it sure doesn't sound the same"

 

"open the flood gates, bring on the discussions, lets talk about it"

 

There is at least one manufacturer  who doesn't know it all and even uses the word Mystery twice. 

 

Hang on the your CD's folks or word up on those that are prepared to sail out in to the big streaming  yonder.

[maximus]

1 hour ago, georgehifi said:

The better CD players have very good multi regulated linear supplies, where as most I know of streamers, have noisy smps in them.

Not my Lumin, quite as a mouse, the better Streamers also have LPS's. Have a listen to an X1 upscaled to 768 or DSD 512, then comeback and tell me how good it sounds.  

[jeromelang]

streaming companies are using a much lower transfer bit-rate to stream those music into your devices.

(often half as low as the original transfer bit-rate of the CD digital format)

 

is that consider a form of "compression"?

 

 

[jeromelang]

The track streamed via Tidal is not as good as the wav file (provided directly by the record label) played through the same streaming device/dac.

 

The transfer bit-rate of the streamed content is more than halved compared against the wav file.

 

Here, I'm using the app on my phone which shows the transfer bit-rate of each of them.

 

Tidal: 671kbps

Wav: 1411kbps

 

 

 

 

 

Edited January 6, 2023 by jeromelang

[jeromelang]

Transfer bit-rate differences.

 

Dire Straits - downloaded wav file (purchased from HD Tracks) vs Tidal

 

Wav: 1411kbps

Tidal: 771kbps

 

The Tidal track is easily discernible by its flatter and smaller soundstaging (especially height).

The download is fuller bodied with more palpable imaging and slam factor. 

 

 

 

 

 

 

Edited January 6, 2023 by jeromelang

[mjs]

27 minutes ago, jeromelang said:

 

 

The transfer bit-rate of the streamed content is more than halved compared against the wav file.

 

 

 

The difference in bite rate is due to the fact that the streamed file is FLAC, rather than WAV, which will have a variable bitrate depending on the amount the original native file has been compressed, which is then uncompressed on playback/streaming. Whereas your CD file is WAV which plays natively at 1411kbps . We can get into all sorts of discussion about whether FLAC is different to playing the original files, but perhaps for another day. Hell, I always rip to AIFF (like WAV).

 

No reason why they couldn't sound the same, subject to being the same original file and the playback chain being competent 

Edited January 6, 2023 by mjs

[Cloth Ears]

1 hour ago, mjs said:

 

...subject to being the same original file and the playback chain being competent 

There's the rub (thanks to Bill)

[Keith_W]

Sorry George, but this is what I got from that video: 

 

- Subjective assertion: Tidal / Qobuz sounds worse than a locally hosted file. 

- Speculation #1: this is because files are compressed at the server and transmitted in compressed form. 

- Assertion #2: Once it gets into your home network it is subject to electrical noise, jitter, etc. which further degrades its sound quality. This can be fixed with an upcoming product. 

- Admission: "I do not know what is happening here". 

 

I note there are a few things missing from this video. 

 

1. No attempt is made to prove that Tidal / Qobuz sound quality is inferior to locally hosted files. I would like to see him (or someone) perform a null test in Audacity to show there is a difference. 

2. He does not mention that lossless file compression that Tidal / Qobuz claims that they do, are able to reproduce the audio waveform exactly as it was prior to compression. 

3. No description of his upcoming magic box or how it works, let alone measurements showing what it does. But i'll give him a pass here, it is an unreleased product so let's wait until it is released before asking for proof that it works. 

 

Unless you already believe that streaming sounds worse than a locally hosted file, I am afraid that I do not find this video to be persuasive at all. 

[georgehifi]

2 hours ago, Keith_W said:

Sorry George, but this is what I got from that video: 

Each to his own.

 

2 hours ago, Keith_W said:

Unless you already believe that streaming sounds worse than a locally hosted file, I am afraid that I do not find this video to be persuasive at all. 

Again each to his own.

 

Cheers George

[Keith_W]

Believe me my intention is not to dismiss the assertion. I am open to the idea that FLAC and Tidal could sound different. I just do not like videos or articles where only assertions are made with no attempt at proof. 

 

Given that McGowan's video only contained assertions, I went around googling for evidence that Tidal and FLAC might be different. I did not find it, but what I found instead was this thread where the OP found that Tidal sounded different to FLAC and Roon, then proceeded to do a null test within the OS (no difference) and from the DAC output. He compared the output of Tidal's own player and Tidal through Roon at the DAC output, and the difference did not null out - suggesting that Tidal's own player implements some DSP (or is simply a bad player) with significantly boosted high frequency making it sound harsh. Streamed FLAC from Tidal (and played through Roon) compared to a purchased FLAC from HDTracks (played through Foobar and Roon) nulled out. 

 

From his post (bear in mind, only one point of evidence) suggests that streamed FLAC from Tidal is identical to a purchased FLAC from HDTracks provided that Tidal's own media player is not used. I have not found evidence to support McGowan's assertion that compression somehow has something to do with it. 

[was_a]

I think when you try to evaluate sound quality on the basis of 'evidence' you're in trouble...

 

 

 

 

 

 

Edited January 6, 2023 by was_a

[stereo coffee]

If streaming services are using known rates of compression, it needs to be looked on, as absolutely the best thing possible. 

 

What needs to be known is the rate at which compression is done, to then apply expansion at exactly the same rate, here is the same idea applied to LP records   

 

[metal beat]

20 minutes ago, stereo coffee said:

If streaming services are using known rates of compression, it needs to be looked on, as absolutely the best thing possible. 

 

What needs to be known is the rate at which compression is done, to then apply expansion at exactly the same rate, here is the same idea applied to LP records   

 

 

Luv this guys graphic equaliser and his settings.  Pumped up bass and treble.

[georgehifi]

7 hours ago, stereo coffee said:

What needs to be known is the rate at which compression is done, to then apply expansion at exactly the same rate

Obvious you have never listened to a DBX Expander or similar, rubbish rubbish rubbish.!!!!!!!!

 

6 hours ago, metal beat said:

Luv this guys graphic equalizer and his settings.  Pumped up bass and treble.

 

Shows what kind of a tool he is, unless it's a  send up. (I couldn't bring myself to press play to find out)

Cheers George

[stereo coffee]

 

IMO Paul is on one hand quite right with everything being said, but being a designer and CEO of a family owned electronics business, needs to be asked, as to his awareness of techniques available to preserve dynamic range with recordings since 1966, and how they may relate to today's streaming service delivery ?  

 

As what is possible when two and two are looked at properly, is IMO far better than the MQA product, which would be truly exciting. 

 

 

 

[bob_m_54]

Dynamic Range compression, and file data compression are two completely different things..