Science shows the real way to listen
Nov 23, 2014 at 11:42 PM Post #2 of 17
I have reservations on the article claiming that Steve Jobs never listened to MP3s, only vinyl.

If true, the epitome of irony?

I purchase music downloaded from iTunes and appreciate it greatly but maybe it's a different experience than streaming massive choice to pick and play from likes of Spotify?

I often use YouTube but tend to have relatively low attention span when using it and flick through music there often not listening to each track fully from beginning to the end.

I think it's natural to give more quality time to music you love enough to purchase, whatever format you choose be it vinyl or whatever digital download format.
 
Nov 24, 2014 at 12:09 AM Post #4 of 17
Oh this'll be fun.
 
Skipping and skimming: Poppy Crum, senior scientist at Dolby Laboratories and consulting professor at Stanford's CCRMA school (Center for Computer Research in Music and Acoustics), spoke with Mic about the psychological effects our digital music habits are having. 
"True love or appreciation for a piece of music ... comes with depth of knowledge of that music," she said. She cited three important factors in creating a genuine experience with a piece of music — "repeated exposure, iterations and intent" — which can be short-circuited in a "taste and go" environment. To her mind, though, the benefits of streaming in terms of access and broad music appreciation far outweigh the potential negative effects of streaming habits on our emotional experience.
 
"Those sorts of heightened emotional responses of pleasure and enjoyment and satisfaction come in a way that is counter to rapid, quick streaming and constant exposure to a lot of different things," Crum said.
 
Skipping and listening inattentively, then, can get in the way of building those connections: "[It] wouldn't be experienced initially, and would bypassed very quickly in a sort of 'taste and go' streaming environment."
 
But that's exactly how we listen now. Recently collected Spotify data illustrates how short our musical attention spans have become. There's only about a 50% chance we'll actually make it to the end of a song. If people are barely listening to a song once all the way through, they're likely not returning to build those emotional connections. If they do, they might not have a foundational experience on which to form them

That's not digital's fault. The medium is only the enabler. It is just as easy, or even easier, to achieve all three of Poppy's factors with digital as it is with analog. There is nothing special about analog that allows these factors to exist, except for the fact that it's less convenient and forces a user to listen a certain way (that is, all the way through an album).
 
The 50% skip data is taken slightly out of context as well. If you check the source link, you'll see that nearly half of the skips (so about 25% of songs) happen within the first 5 seconds, and 2/3rds (35% of all songs) happen within the first 30 seconds. It's not like people are listening to a song they like, get bored, and skip to the next one half way through. Most skips appear to be the result of using a preset station or random playback, and a song the user doesn't like or hasn't heard comes on.
 
BTW I'm all digital, raised on digital, though I avoid streaming, and I almost never skip songs unless I'm really not in the mood. I listen to every album I add to my playlist first. And I refuse to delete songs from albums, I'm very particular about having whole albums. Meanwhile my mom, who grew up in the 60s and tolerated my dad and his turntable through the 80s, will only cherry-pick the hits off CDs. It's not a matter of what medium we're used to, or what era we were raised, it's a matter of listening habits and level of passion. That's the problem with statistics
wink.gif

 
Many music professionals have also discussed this lack of connection, and they blame the dwindling quality of audio files for it. When record companies digitally convert recorded music, which consumes a ton of data in its original form, they turn it into the much smaller MP3 format. 
But this compressing process strips about 91% of the actual musical data and fills in the gaps using algorithms. The volume is then jacked up to make up for this lack of distinctiveness, and the resulting waveform is barely recognizable. Not only that, it can actually exhaust your ears to listen to it. It ends up looking like a solid brick of noise, as the following portion of the infographic "A Visual History of Loudness," created by designer Christopher Clark, shows.
 
Bob Ludwig, a record mastering engineer, believes this is one of the chief reasons people don't engage with albums as deeply anymore. "When you're through listening to a whole album of this highly compressed music, your ear is fatigued," he told NPR. "You may have enjoyed the music but you don't really feel like going back and listening to it again."
 
Research shows that musical quality has a huge effect on emotional response. A recent study performed by audio researchers at DTS divided a group of listeners into two groups — one that watched a video accompanied by standard stereo 96-kbps sound (Spotify's default audio setting) and the other group listened in 256-kbps audio format. The responses in the brains of the group listening with the 256-kbps audio were 14% more powerful on metrics measuring memory creation and 66% higher on pleasure responses. And this was just 96 to 256 kbps. 
 
Vinyl records are estimated to play at a whopping 1000 kbps. Music might not just have lost its revenue when it switched to digital; it may have lost its emotional power too.

Yikes, here's where the typical misunderstandings of digital audio show up. Lossy compression is a completely separate issue from dynamic compression, mentioned in this first paragraph in one breath. And, again, neither is some intrinsic property of digital audio. Lossless digital audio exists, as I'm sure every listener on this forum knows by now. It's a means of distribution, used in streaming and download services to reduce file size and server load. Dynamic compression, meanwhile, is a part of the mastering process. Like our listening habits, digital as a medium can only be blamed as an enabler; the use of dynamic compression is made possible by digital's ability to shape the sound beyond the physical means of analog. In the right hands it can actually be a powerful tool, but these days is mishandled.
 
256 kbps scored 66% higher in please than 96 kbps? That's reasonable, 96 kbps is incredibly low and bound to have audible artifacts. But "just" 96 to 256? The author is falling into the trap of believing more bits = better sound, a misconception. Lossy codecs are very good at what they do, and I'm willing to bet that 256 kbps is transparent for 95% of listeners. The "1000 kbps" of analog isn't going to score 66% higher again in pleasure, because the audible differences between 256 and 1000 in this case are no doubt much smaller than 96 to 256. Also, hint: Vinyl doesn't have bit rate because it's analog, but it can be fully captured by a 1411 kbps lossless 16/44.1 digital file, then losslessly compressed to around 1000 kbps. Hear that? A digital file can perfectly capture an analog recording, as in play it back exactly. Why is the medium to blame again?
 
Correct me if I'm wrong, but isn't the music industry making more now than in the vinyl era? What's this talk of revenue? That's the only mention of the word in the article.
 
I won't quote the rest because it's just a circular argument.
 
Huh, I expected it to be longer than that.
 
In summary, because the advances of digital audio, downloads and streaming allow for greater flexibility in listening, there is an increasing trend for listeners to take a passive approach to audio consumption. Therefore, digital is inherently evil and we should force everyone to instead use an antiquated technology that significantly impedes their freedom to listen how they like because there should only be one way to listen.
 
Nov 24, 2014 at 12:10 AM Post #5 of 17
From the article:
Vinyl records are estimated to play at a whopping 1000 kbps. Music might not just have lost its revenue when it switched to digital; it may have lost its emotional power too.
 
Without knowing enough to mention that even more than vinyl, CD has 1411 kbps.  But then that would show their contorted conclusions and logic didn't bear scrutiny. 
 
Nov 24, 2014 at 12:13 AM Post #6 of 17
From the article:
Vinyl records are estimated to play at a whopping 1000 kbps. Music might not just have lost its revenue when it switched to digital; it may have lost its emotional power too.

Without knowing enough to mention that even more than vinyl, CD has 1411 kbps.  But then that would show their contorted conclusions and logic didn't bear scrutiny. 


And their source for that 1000 Kbps is from a Yahoo Answers post. :D

se
 
Nov 24, 2014 at 12:54 AM Post #12 of 17
Wonder what source he used to conclude that data compression was the same as dynamic range compression? Wahoo Answets? :D

se


I'm not going to pretend to understand the difference between data and dynamic range compression but understand by your wahoo answers comment that the author of article was grossly incorrect in concluding they were the same :)

Edit: maybe this qualifies me as a journalist for MusicMic? :)
 
Nov 24, 2014 at 1:22 AM Post #13 of 17
It's just a badly written article really. As already stated above, the author seems to be getting a bit confused. I'm surprised how often dynamic range compression and data compression are mixed up.
 
At least there are plenty of comments pointing out the confusion.
 
Nov 24, 2014 at 2:16 AM Post #14 of 17
I'm not going to pretend to understand the difference between data and dynamic range compression but understand by your wahoo answers comment that the author of article was grossly incorrect in concluding they were the same
smily_headphones1.gif


Edit: maybe this qualifies me as a journalist for MusicMic?
smily_headphones1.gif


Data compression is taking a 44.1khz CD and turning it into an MP3.  The total bits of the CD is 1.411 mbps.  Data compression will throw away most of the bits using an algorithm which has a human hearing model.  The model will estimate which info we actually cannot hear whether it is there or not.  That we can't hear it will throw away.  Such models are not perfect so if enough data is thrown away you can hear the difference.  So a CD starts as 1.411 megabits per second.  It can be turned into an MP3 of .128 megabits per second. More than 90% of the bits are thrown away. 
 
Dynamic range compression is altogether different, and has nothing directly to do with data compression (aka lossy encoding formats).  Let us say you have a recording and the difference between the loudest and quietest parts is 60 db.  You do 2 to 1 compression, and now the difference is only 30 db.  The difference between loud levels and quiet levels have been squashed together.  More extreme compression like 10 to 1 would turn 60 db dynamic range into only 6 db.  A continuously loud bit of music which will lose much of the nuance and artistry by being squashed so much.  
 
Too much of either compression can be bad sounding.  But one has nothing to do with the other.  That fact is quite often confused by journalists.  
 
Nov 24, 2014 at 2:45 AM Post #15 of 17
Data compression is taking a 44.1khz CD and turning it into an MP3.  The total bits of the CD is 1.411 mbps.  Data compression will throw away most of the bits using an algorithm which has a human hearing model.  The model will estimate which info we actually cannot hear whether it is there or not.  That we can't hear it will throw away.  Such models are not perfect so if enough data is thrown away you can hear the difference.  So a CD starts as 1.411 megabits per second.  It can be turned into an MP3 of .128 megabits per second. More than 90% of the bits are thrown away. 

Dynamic range compression is altogether different, and has nothing directly to do with data compression (aka lossy encoding formats).  Let us say you have a recording and the difference between the loudest and quietest parts is 60 db.  You do 2 to 1 compression, and now the difference is only 30 db.  The difference between loud levels and quiet levels have been squashed together.  More extreme compression like 10 to 1 would turn 60 db dynamic range into only 6 db.  A continuously loud bit of music which will lose much of the nuance and artistry by being squashed so much.  

Too much of either compression can be bad sounding.  But one has nothing to do with the other.  That fact is quite often confused by journalists.  


Thank you for the detailed explanation :)
 

Users who are viewing this thread

Back
Top