iTunes vs Spotify vs the original WAV file

[MajorFubar]

I recently remixed and remastered a piece of music I'd recorded some years ago and I've had it uploaded to iTunes via a distributor. It's available on Google Play and on Spotify too. I won't post a link to it because I'm sure this forum isn't here to provide free publicity to independent artists.

But more interestingly, what it has allowed me to do is compare the quality of the Spotify/iTunes versions with my original 16bit / 44.1kHz master.

(Yes, I downloaded my own track on iTunes, just for the sake of comparison.)

Verdict: the difference is not huge. When you remove all other variables in the equation, chiefly being which master has been used, the difference is minimal. Pushed to split hairs, the download/streaming versions lack a tiny amount of bite and spaciousness in the upper-mid / low treble, and the bass isn't quite as crisp and taut, nor do they have quite the same slam. But this is only evident when I'm deliberably looking for differences. Which of the two versions sounds most like the original? The iTunes download. But not by a country mile.

 

[JMac]

Very interesting to hear that. You always expect such providers to compress the hell out of a file.

 

[MajorFubar]

See I thought this would actually generate some considered interest, written as it is from the unusual perspective of an amateur musician who has created the complete end-to-end product.

Ah well.

 

[davedotco]

What cables did you use in the studio....... :rofl:

Actually I was wondering what was the resolution of the download.....?

I have heard some very good things about Apple's variable bit rate technology but since I do not buy from iTunes i really don't know so there is little that I can say.

I am hoping for an affordable streaming solution from Apple, based on Beats Music. Something just a little better than Spotify would be nice.

FWIW. I find Spotify perfectly satisfactory in comparative tests but not quite as engaging as it could be long term. I am hoping a slightly better option might fix that.

 

[Overdose]

You could try the Foobar ABX and see how they fare, but your findings don't surprise me.

iTunes has a good reputation for quality downloads. Did you use the Mastered For iTunes process?

Also, what compression was used for the upload, or did you send the WAV? Just wondered if the file was compressed first and then transcoded, which might account for variables.

 

[Overdose]

I have heard some very good things about Apple's variable bit rate technology but since I do not buy from iTunes i really don't know so there is little that I can say.

VBR AAC files are excellent and at 256Kbps, quite indistinguishable from the master they came from, I find.

 

[Vladimir]

To a normal person differences are small. However, to an audiophile they are big as an elephant in a bathroom.

 

[MajorFubar]

The uploaded file was 16-bit / 44.1kHz.

What cables did you use in the studio....... :rofl:

Actually I was wondering what was the resolution of the download.....?

lol! Cables? Just don't go there. Resolution...good question. It's iTunes Plus, so 256K AAC I think. My Spotify Premium is 320k OGG as far as I'm aware.

 

[MajorFubar]

Did you use the Mastered For iTunes process?

Sort of. I have the AURoundTripAAC Audio Unit in Logic Pro which brings your attention to any issues that may arise during AAC transcoding, but I haven't downloaded the full Mastering Tools toolkit.

 

[Overdose]

To a normal person differences are small. However, to an audiophile they are big as an elephant in a bathroom.

Absolutely. Whether they exist or not.

 

[davedotco]

The uploaded file was 16-bit / 44.1kHz.

What cables did you use in the studio....... :rofl:

Actually I was wondering what was the resolution of the download.....?

lol! Cables? Just don't go there. Resolution...good question. It's iTunes Plus, so 256K AAC I think. My Spotify Premium is 320k OGG as far as I'm aware.

I really do not know my way around the itunes store at all. Can you chose the quality of the download? Low res, 256 AAC, VBR, Mastered for iTunes etc.

I use Spotify almost exclusively and accept the quality as it is, be nice to have some better options though........!

 

[Overdose]

Did you use the Mastered For iTunes process?

Sort of. I have the AURoundTripAAC Audio Unit in Logic Pro which brings your attention to any issues that may arise during AAC transcoding, but I haven't downloaded the full Mastering Tools toolkit.

OK, thanks. Was thinking that the transcoding post compression would be the most likely source of any minor discrepencies, but as you say, effectively imperceptible anyway.

 

[davedotco]

To a normal person differences are small. However, to an audiophile they are big as an elephant in a bathroom.

Absolutely. Whether they exist or not.

I have a second bathroom, I keep the elephant in there.

Facing the door of course....... 8)

 

[MajorFubar]

I really do not know my way around the itunes store at all. Can you chose the quality of the download?

No. Everything you download is 256k AAC as far as I'm aware, though iTunes Match and iTunes Radio stream at less. Not sure at what bitrate off the top of my head, probably 160k AAC. Mastered for iTunes is sort of a mastering discipline. There are guidelines and tools you can run to make sure your master sounds as good as possible as an AAC download.

The RoundTripAAC Audio Unit in Logic Pro is really cool though. Not only does it show you any technical problems which may arise from transcoding to AAC, it'll hit you with an ABX Listening Test.

 

[davedotco]

I really do not know my way around the itunes store at all. Can you chose the quality of the download?

No. Everything you download is 256k AAC as far as I'm aware, though iTunes Match and iTunes Radio stream at less. Not sure at what bitrate off the top of my head, probably 160k AAC. Mastered for iTunes is sort of a mastering discipline. There are guidelines and tools you can run to make sure your master sounds as good as possible as an AAC download.

The RoundTripAAC Audio Unit in Logic Pro is really cool though. Not only does it show you any technical problems which may arise from transcoding to AAC, it'll hit you with an ABX Listening Test.

Thanks, this is all new to me, really must do some reading up...... :cheers:

 

[MajorFubar]

NP anytime :cheers:

 

[the record spot]

To a normal person differences are small. However, to an audiophile they are big as an elephant in a bathroom.

You should try that in the Steve Hoffman forum. It'd take over the whole house there...!

 

[Vladimir]

I don't own a turntable or tubes and I talk too much. I wouldn't last a week in there.

 

[tonky]

In a decent hifi set up the differences between 256 mb/s and the same music in wav can be heard - hopefuly not to contentious a statement.

tonky

 

[Neptune_Twilight]

Despite having over 800 albums ripped to .flac I do tend to listen to Spotify a fair amount of the time, there are differences but I've found a lot of tracks on Spotify that would mean buying an entire often out of print CD for just a single track & for use in the background or working at home Spotify is listenable - If Spotify ever does streaming as opposed to downloadable .flac quality that would be great as I would never buy music again even if there was a price hike - I've got 160 megbit download ability on Virgin fibre & it's a pity I can stream HQ music.

 

[BenLaw]

So any chance of you doing an ABX on these major?

 

[manicm]

Not going to get into what sounds better, but just want to offer my niche experience: my CD drives seemed to make a difference but with settings you wouldn't expect:

My old Dell M1530 XPS laptop, 6 years old, gave up the ghost earlier this year and I had just begun ripping all my CDs in WAV with EAC - but configured for 'quality', not 'speed'. The results were good. It had a nice slot-loading drive.

With my new laptop, a mid-gaming MSI device, with the same EAC settings the results seemed slightly worse and worringly, very slow. I then configured it for 'speed', and lo and behold, I found the rips excellent sounding, and of-course fast. This was totally unexpected, and perhaps EAC just did not know how to deal with this specific drive, and I didn't want to delve into the deeper settings, but anyway I'm getting satisfaction.

BTW the latest version of EAC is excellent if you just want to rip WAVs like I do, and it's free. It also handles album art very well.

I'm ripping all my CDs with the view of locking them up eventually, and also in preparation for a future streaming system.

 

[drummerman]

Interesting and thanks for the comparison.

I use Spotify Premium and rip my cd's using dbpoweramp lossless Flac with Musicbee as player.

Whilst there is a slight decrease in soundquality I find Spotify easy to listen to, so much so that the issue of SQ never usually enters my head. Ogg Vorbis has a good reputation for Sound quality and I think this manifests itself with Spotify.

Would I pay more for higher quality playback? Perhaps but not to much more. I am not sure if there is a viable alternative to spotify but I love the interface and selection.

regards

 

[Broner]

Would I pay more for higher quality playback? Perhaps but not to much more. I am not sure if there is a viable alternative to spotify but I love the interface and selection.

Could very well be the case that either Spotify, Google Play Music or Deezer offers higher quality playback in the future. Not sure if I would find it relevant, but I suppose relevance didn't stop the megapixels race.

 

[MajorFubar]

So any chance of you doing an ABX on these major?

probably could be arranged but I don't think it would prove anything useful because the differences seem really small to start with, especially the iTunes version. The differences are mainly limited to upper-mid airiness and snap, and a slightly tauter bass. The iTunes versions seems fractionally closer, but I wouldn't like it to be a $64,000 question. In fact so small are the differences that despite the quality of my EB2s I can only hear any difference while critically listening on my £200 Sennheiser headphones. I have the significant advantage of having composed, performed and produced the music in the first instance so I'm proactively listening for things I know are there, and even armed with that sizeable advantage, the differences seem really so small that I'm not sure I would consistently pass an ABX test with a statistically-relevant accuracy.