Web Rip vs CD Rip

[One and a half]

One of the many reasons not listening to Tidal was that the music sounded different compared to the Cd rip and played locally.

 

I came across a Web rip of a very familiar album that I've owned since 1987, it's Thom Rotella Band self titled CD on DMP. DMP were pioneers of digital recordings back in the day, the Roon DR range on this particular album is 18. Tracks feature percussion with all kinds of hidden triangles, shakers, and the drum kicks are as about 'real' as there is. The album comes out when there's a new piece of equipment, or experimenting with interconnects to reset the reference and to find any new gems.

 

The Web Rip when listened to on speakers, strange, something was missing and something added.  The attack of the drums is missing, and sounds mush. The analyze audio from Jriver was not that conclusive, since the playback volumes were different. Opened sound forge and measured the frequency analysis, without tinkering with too many settings.

Well, there's some differences.

 

Web Rip, Patti Cake 30s sample

 

 

CD Rip Patti Cake 30s sample

 

On the web rip is DC? Going years back now, Audacity had a habit of creating oddities, was this one of them? Being that as it may, the upper ranges below 22kHz are very different, with 19k has little worth. Either the source or audacity (?) aren't working to the full fidelity, vindicating my view streaming media is great for cars, background listening and exploring new music but for critical listening, not to be taken seriously.

01 Patti Cake CD Rip 30s.wav 01 Patti Cake Web Rip 30s.wav

[sandyk]

On ‎4‎/‎4‎/‎2020 at 10:36 AM, One and a half said:

Either the source or audacity (?) aren't working to the full fidelity, vindicating my view streaming media is great for cars, background listening and exploring new music but for critical listening, not to be taken seriously.

 

 

 The Web rip sounds pathetic in comparison with the CD rip !

[opus101]

Just from a quick eyeballing in Audacity - the channels are swapped and one of the two is phase inverted.

[One and a half]

15 hours ago, opus101 said:

Just from a quick eyeballing in Audacity - the channels are swapped and one of the two is phase inverted.

The phase to inversion was me, it was a few days in between checking, and I normally don’t save any changes, maybe SF14 does. When I’m back home can redo. 
the channel swap is the web rip.

 

found another will transmit

[One and a half]

Update

 

Extracted an original Web rip FLAC converted to WAV for analysis of the same Patti Cake track. I thought I'd saved a null exercise by mistake, but under audacity and with an expanded zoom, (should have used this originally), the phase inversion and channel swap is clearly evident, thanks to @opus101 for pointing this out.

 

Web Rip

 

 

CD Rip

 

 

The differences are clear in the first 1.5s when the music starts.

[The Computer Audiophile]

On 4/3/2020 at 6:36 PM, One and a half said:

vindicating my view streaming media is great for cars, background listening and exploring new music but for critical listening, not to be taken seriously.

With a sample size of one, I wouldn’t be so quick to dismiss streaming. 
 

That said, very interesting details. Thanks for sharing. 

[One and a half]

22 minutes ago, The Computer Audiophile said:

With a sample size of one, I wouldn’t be so quick to dismiss streaming. 
 

That said, very interesting details. Thanks for sharing. 

I've tried Tidal for a few weeks, Qobuz regional restrictions and billing prevent listening and found something 'missing' compare to a local CD, rotating or from a digital file.  Drums sounded like a paper bag being hit as an extreme illustration, but there was no life. 

Perhaps locally, there's too much noise on the network, and/or I could (at the time) use chrome to browse and select music. I do have Roon that can play Tidal, from what I read on the Roon forums, all is not that well either, mainly from a management type of view, data downloads, playback as examples.

 

I will keep looking for more samples like the one above.

[cjf]

Breaking out my shovel to dig this back up based on my own recent experience.

 

I've just started a Trial of Qobuzz and this is my 2nd time trying streaming after a long hiatus with first trying Tidal. Anyway, I agree 100%, so far, with the original poster.

 

I'm not even finding it subtle the difference in SQ between full CD Rip of Redbook quality Albums vs what looks to be the same version of the albums I've tried so far found on QoBuzz. The whole soundstage is shrunk, the level of output in terms of volume is reduced and the bass is flabby and not accurate sounding at all.

 

Its not likely anything to do with my Internet connection, its speed nor other network equipment involved. Basically the outcome is the same as my previous attempt to dip my toe into the streaming waters. For any serious listening its the wrong tool for the job IMO and IME thus far but for discovery purposes its perfectly great and I have no complaints in that regard. I've certainly wasted more money on a given day on dumb stuff than it cost to discover music you don't currently own yet but were always curious about trying.

 

So in summary I think it certainly sounds better than any .mp3 version of a given album but not at the same level as a local CD rip or even a straight up download of an album from the same company. Something is being lost in translation from point A to B. Honestly it sounds very close to what I hear when comparing wireless to wired ETH connections while playing music.

[The Computer Audiophile]

1 hour ago, cjf said:

Breaking out my shovel to dig this back up based on my own recent experience.

 

I've just started a Trial of Qobuzz and this is my 2nd time trying streaming after a long hiatus with first trying Tidal. Anyway, I agree 100%, so far, with the original poster.

 

I'm not even finding it subtle the difference in SQ between full CD Rip of Redbook quality Albums vs what looks to be the same version of the albums I've tried so far found on QoBuzz. The whole soundstage is shrunk, the level of output in terms of volume is reduced and the bass is flabby and not accurate sounding at all.

 

Its not likely anything to do with my Internet connection, its speed nor other network equipment involved. Basically the outcome is the same as my previous attempt to dip my toe into the streaming waters. For any serious listening its the wrong tool for the job IMO and IME thus far but for discovery purposes its perfectly great and I have no complaints in that regard. I've certainly wasted more money on a given day on dumb stuff than it cost to discover music you don't currently own yet but were always curious about trying.

 

So in summary I think it certainly sounds better than any .mp3 version of a given album but not at the same level as a local CD rip or even a straight up download of an album from the same company. Something is being lost in translation from point A to B. Honestly it sounds very close to what I hear when comparing wireless to wired ETH connections while playing music.

Certainly not your system. I love that system. It must sound sublime. 

[BigWilliam]

16 hours ago, cjf said:

Breaking out my shovel to dig this back up based on my own recent experience.

 

I've just started a Trial of Qobuzz and this is my 2nd time trying streaming after a long hiatus with first trying Tidal. Anyway, I agree 100%, so far, with the original poster.

 

I'm not even finding it subtle the difference in SQ between full CD Rip of Redbook quality Albums vs what looks to be the same version of the albums I've tried so far found on QoBuzz. The whole soundstage is shrunk, the level of output in terms of volume is reduced and the bass is flabby and not accurate sounding at all.

 

Its not likely anything to do with my Internet connection, its speed nor other network equipment involved. Basically the outcome is the same as my previous attempt to dip my toe into the streaming waters. For any serious listening its the wrong tool for the job IMO and IME thus far but for discovery purposes its perfectly great and I have no complaints in that regard. I've certainly wasted more money on a given day on dumb stuff than it cost to discover music you don't currently own yet but were always curious about trying.

 

So in summary I think it certainly sounds better than any .mp3 version of a given album but not at the same level as a local CD rip or even a straight up download of an album from the same company. Something is being lost in translation from point A to B. Honestly it sounds very close to what I hear when comparing wireless to wired ETH connections while playing music.

 

Huh that's interesting. I have many downloaded tracks but the only time i do stream it's just through the web browser through windows. I just assumed all difference I was hearing was from not being on a better output driver and my receiver's dac.

 

I'm not an audiophile by this forums standards or an expert. But I do play video games and sometimes when trying to connect to a server far away you get noticeable "packet loss". Quick google reveals this little study from some Muslim guys outta Europe maybe/ First link on google. Random people notice at very low bit rates at 5% packet loss

 

https://www.researchgate.net/publication/335997270_Effect_of_Packet_Loss_and_Reorder_on_Quality_of_Audio_Streaming

 

So i would imagine it's not your network, but it is some ones network that's messing up the playback.

 

Obviously assuming the steaming service isn't purposely throttling the bitrate in some way.

 

edit - there is some sort of error correction going on here too. Above my pay grade. I'm sure these services have better packet loss protection than VLC.

 

https://inst.eecs.berkeley.edu//~ee290t/sp04/lectures/packet_loss_recov_paper11.pdf

 

 

 

 

 

 

 

 

 

[cjf]

Thanks for the kind words Chris and likewise to your setup as well.

 

Just to add two specific examples. A few months back I purchased/downloaded Tool's new album Fear Inoculum in 24/96 .wav format from QoBuzz. IMO this album is the best one the band has made in terms of musical content and SQ and I've come to know it very well since that time (probably have heard it upwards of 20 times now). Just for giggles I cued up Track #2 "Pneuma" from my Local copy version and then I also cued up the same song at the same Bit/Sample Rate directly from the QoBuzz streaming catalog (all via Roon) and had them both in my playback queue in Roon.

 

I then proceeded to conduct a caveman version of a blind test against the two songs. With the songs in the queue I repeatedly tapped the Shuffle button at least 10 or so times so that I didn't know which one was playing. After the song played twice I proceeded to tap the Shuffle button again 10 or so more times to mix it up again. Long story short, I listened to the song 4 times back to back. After each pair of playback sessions I picked which version I preferred. In short, I picked the Local downloaded version both times correctly as sounding noticeably better. Not the most scientific test but close enough for my purposes.

 

The same test was performed again using another Album/Track I know very well that I own the download of which is Led Zeppelin II (Track #1 Whole Lotta Love) in 24/96 .wav. During this test the Local copy was picked again both times as the preferred version in the same blind fashion over the streamed version.

 

To be clear, I am in no way blaming QoBuzz for their services but simply saying that I feel there is something about the streamed version of the songs that is quite different sounding than the Local downloaded version. I may still choose to purchase a monthly subscription from QoBuzz despite this finding because I do find it nice to be able to listen to pretty much anything at any time I wish even if its just for background music while doing other things around the house.

 

One thing that would be interesting to try (if it were an option while using Roon as the host system to connect to QoBuzz) would be to pull the tracks down in "Offline Mode" to listen to instead of streaming them from the Cloud directly. I know this option exists for the QoBuzz App but I didn't see it as an option within Roon. I may have just missed it though. In theory listening to the albums in offline mode should eliminate the differences heard in SQ I would think but who knows.

 

 

[The Computer Audiophile]

42 minutes ago, cjf said:

Thanks for the kind words Chris and likewise to your setup as well.

 

Just to add two specific examples. A few months back I purchased/downloaded Tool's new album Fear Inoculum in 24/96 .wav format from QoBuzz. IMO this album is the best one the band has made in terms of musical content and SQ and I've come to know it very well since that time (probably have heard it upwards of 20 times now). Just for giggles I cued up Track #2 "Pneuma" from my Local copy version and then I also cued up the same song at the same Bit/Sample Rate directly from the QoBuzz streaming catalog (all via Roon) and had them both in my playback queue in Roon.

 

I then proceeded to conduct a caveman version of a blind test against the two songs. With the songs in the queue I repeatedly tapped the Shuffle button at least 10 or so times so that I didn't know which one was playing. After the song played twice I proceeded to tap the Shuffle button again 10 or so more times to mix it up again. Long story short, I listened to the song 4 times back to back. After each pair of playback sessions I picked which version I preferred. In short, I picked the Local downloaded version both times correctly as sounding noticeably better. Not the most scientific test but close enough for my purposes.

 

The same test was performed again using another Album/Track I know very well that I own the download of which is Led Zeppelin II (Track #1 Whole Lotta Love) in 24/96 .wav. During this test the Local copy was picked again both times as the preferred version in the same blind fashion over the streamed version.

 

To be clear, I am in no way blaming QoBuzz for their services but simply saying that I feel there is something about the streamed version of the songs that is quite different sounding than the Local downloaded version. I may still choose to purchase a monthly subscription from QoBuzz despite this finding because I do find it nice to be able to listen to pretty much anything at any time I wish even if its just for background music while doing other things around the house.

 

One thing that would be interesting to try (if it were an option while using Roon as the host system to connect to QoBuzz) would be to pull the tracks down in "Offline Mode" to listen to instead of streaming them from the Cloud directly. I know this option exists for the QoBuzz App but I didn't see it as an option within Roon. I may have just missed it though. In theory listening to the albums in offline mode should eliminate the differences heard in SQ I would think but who knows.

 

 

This is really interesting. It’s what all of us would do to test. It’s pretty reasonable. 
 

The results baffle me a bit as I don’t know why you hear a difference. Perhaps I need to purchase this album and run the same tests. 

[DuckToller]

Just now, The Computer Audiophile said:

This is really interesting. It’s what all of us would do to test. It’s pretty reasonable. 
 

The results baffle me a bit as I don’t know why you hear a difference. Perhaps I need to purchase this album and run the same tests. 

+1
This album is very special from it soundscape, mastered by Bob Ludwig, I'd guess it is better with the Wilsons than the RAAL because it needs space to excel ;-)
IMHO, Chris is quite correct with his findings, it sounds very good via Qobut but really special from a local source.
Happy XMAS to both of you, Chris & Chris ;-)

[firedog]

22 hours ago, The Computer Audiophile said:

This is really interesting. It’s what all of us would do to test. It’s pretty reasonable. 
 

The results baffle me a bit as I don’t know why you hear a difference. Perhaps I need to purchase this album and run the same tests. 

The streaming, download, and physical  versions are not necessarily the same - even when they seem to be. The hi-res ripped and downloaded version of the White Album remix don't sound the same, even though ostenibly they are both `'the 24/96 version". Apparently slightly different masterings. Streamed version sounds like the downoad, not the rip. 

[cjf]

2 hours ago, firedog said:

The streaming, download, and physical  versions are not necessarily the same - even when they seem to be. The hi-res ripped and downloaded version of the White Album remix don't sound the same, even though ostenibly they are both `'the 24/96 version". Apparently slightly different masterings. Streamed version sounds like the downoad, not the rip. 

In the Tool album example I gave there is only one version available on QoBuzz. I suppose its possible the stream could be a different version but not sure why they would bother having a different version for download vs streamed in that case.

 

The playback buffering scheme is probably different though when pointing to a Local download on your NAS vs the same songs coming from the Cloud which could be one factor at play.

 

There are ways to capture and analyze both versions using other software which I wont explain how to accomplish here other than to say that both spectrum plots could be exported to a .txt file and compared using Excel. I'll let those interested in that more scientific analysis find out how to do so on there own.