Has the Transcoding Problem Worsened Thanks to YouTube?

If you’re a hardcore rock fan, the name “Deadmau5” might not really ring a bell – unless you know someone who is familiar with the “rave” scene. One of the bigger names in House music had some interesting things to say about how much quality is lost on YouTube even though it’s labelled as “High Quality”. To say he isn’t happy about how much quality is lost is an understatement.

Note: This is an article I wrote that was published elsewhere first. It has been republished here for archival purposes

For some time now, YouTube has had a feature on many videos that let you see videos “in HD”. Of course, this insinuates that the video will feature the highest quality sound you can hope for. The reality is that the quality of the video does depend on the original quality format the uploader has. If the uploader has a low quality video, YouTube isn’t necessarily going to magically turn it into “HD”.

While discussing an EU commissioner wanting to overhaul regulations related to downloading, we noted a British study that suggested that file-sharing is actually going down in popularity. The source that is gaining on the back of file-sharing’s decrease in popularity? Online streaming. Many are, in fact, getting their music from sites like MySpace and YouTube. Long before the release of the study that pointed to an increase in popularity of streaming, we wrote a guide on how to rip music from MySpace. There’s plenty of ways to get music from YouTube as a simple Google search reveals countless online YouTube stream rippers.

This phenomenon of users getting their music from streamed sources has caught the attention of a number of artists. Recently, this included an artist by the name “Deadmau5” (pronounced ‘Dead mouse’). For the uninitiated, Deadmau5 is a famous House music producer from Canada. If you’re even somewhat into House music (or many forms of electronica for that matter), there’s a very good chance you’ve heard of this artist. In terms of popularity in a given genre, you could consider him the Linkin Park of House music only with significantly less lyrics.

“Quite frankly,” Deadmau5 wrote, “I kinda feel a bit bummed out that i would create something exclusive for my audiences to be “ripped” from some streaming media [site] as is… moreso when the poster of the media would advertise or title it as “HD” or “High Quality” before it’s release date.”

A little bit of background. Many artists such as Deadmau5, during a live show, beatmatches music together so there’s essentially a constant amount of music being played even though several different songs are played during a live show. Often, larger artists play their music long before their release dates to help promote the new song and increase popularity. Of course, similar styles of performing live transcends to radio as well. This is why some electronica music you hear seems to cut in and out at the end or get cut off completely. It’s called a set rip. There’s still that beatmatching and, consequently, that songs beginning (mostly drum kicks) and ending (also mostly drum kicks) are more or less cut off.

“In fact, Nothing makes me vomit in my mouth more than listening to anything short of a 16bit 44.1kHz Wav / Aiff file… perhaps it’s just the audiophile in me… but i am actually witnessing and realizing for the first time in my life something that i have created that has been reduced to some regurgetated re-re-re-re-recording of a set and dubbed “high quality”… quite frankly im a little embarassed.” Deadmau5 continues.

A WAV file is essentially an uncompressed sound file. When someone encodes it to MP3, the “highs” and “lows” start being lost in an effort to make the file smaller. A good way to demonstrate to yourself what the sound differences are between a high quality and a low quality version of an MP3 is, listened to a song that’s at 320KBps, then compare that to the same song that’s in an MP3 format of 128KBPs. If the differences are seemingly minimal, pay attention to the high hats or cymbals. You may notice that the hats get a little scratchy or distorted as quality is reduced. If you’re listening on, say, a subwoofer system, listen to how heavy the bass is as well.

Admittedly, when it comes to the really high quality stuff, I don’t personally notice the differences. The differences between a 32bit WAV file and a 16bit WAV file is completely inaudible to me. Unfortunately, I don’t own a $50,000 sound system to put my ears to the ultimate test. But if you are wondering, a 16bit WAV file is the quality of a standard CD without any compression (not even to save space as FLAC does without losing any quality). Of course, the difference between 16bit and 32bit isn’t even close to what Deadmau5 is getting at.

“So,” Deadmau5 explains, “situation happens… some guy records the set @ 16 bit 44.1kHz (usually the case) you’ve got my 24 bit version coming out of my DAC (Digital to Analog Converter) then in to someone elses ADC (Analog to Digital Converter) which is only as good as the quality of the ADC.. so already the quality of the music has been compromised, and you havent even gotten it yet.”

“Then, whoever records a direct line from the set then encodes that to 320kpbs MAX or 128 kbps mp3 for media streaming ease. this part is a MAAASSSIIIVEEE degradation of the original signal… and has been mulched to all bloody hell. So…. then what typically happens is yet ANOTHER DAC / ADC conversion…. wheres someone rips the audio from said streaming media outlet only to puke it back onto youtube… which is 128Kbps max in “HD” mode (which is a VERY far cry from HD if you ask me… total abuse of terminology there) shoulda [just] called it ABBDBNR….”A Bit Better Definition But Not Really”. fun times! so… to break it down… lets take a quick look again…”

“Production / Final Master
↔
degraded to 24 bits for live use
↔
degraded to 16 bits for commerical release / distribution
↔
degraded to 16 bits (less accurately) when recorded off a live feed from mixer via DAC / ADC
↔
degraded to 128kbps / 320kbps when published for streaming media outlets
↔
degraded to 128kbps / 320kbps when ripped from streaming media outlet
↔
degraded to 64kbps / 128kbps
=
a very crappy 6th generation copy that magically gets called “HIGH QUALITY VERSION!!!111ONE!!” on youtube…. you figure it out.”

Transcoding has been a problem in the file-sharing community for years, long before the streaming revolution. The situation then was typically someone finding a 128kbps MP3, then re-encoding it into a 320kbps mp3, either thinking that they have a higher quality version of the song or trying to dupe others into believing it’s higher quality. That was typically going along the lines of, “producer quality, degraded to 16bit-like quality for CD, ripped and encoded into a low quality MP3, uncompressed and encoded again to 320kbps”. That’s a 4th generation version, not even a 6th generation version as described by Deadmau5.

Additionally, thanks to standards in streaming and YouTube, the quality is forced to be of a certain level of (lower) quality, so the amount of quality lost is even greater because you lose more quality in each step and that quality doesn’t come back once it’s lost once.

Deadmau5 concludes, “I just think it’s time for us listeners to listen smarter, to educate eachother a little more, give the audiable world out there a bit of a quality control attitude, LISTEN UP folks… our hearing is really a fucking complicated and beautiful thing!!! Why not use it well? It’s like owning the most amazing sportscar and not a drop of gasoline on the planet…. Sometimes when i stick my ears out there, im really starting to fear that even the terminologic use of the words “high quality” is slowing substandardizing itself into an inaudiable heap of nonsense. If my message reaches at least one person, and benefits them in any way, i will die a happy mouse.”

There’s many ways to look at this from an artist standpoint. A positive way to spin the stream ripping activities is to say that if you buy the music, the quality is much greater. Consider it an extra incentive for music fans to legally pay for music for example. We’re not necessarily talking about users who are higher up on the file-sharing food chain here most of the time. Chances are, the users who are listening to ripped streaming music aren’t members of 6 different private sites as the quality difference is monumental between streamed music and, say, FLAC – though people higher up in the file-sharing food-chain are much more encouraged to buy what they like anyway.

Another way one can look at this is the tried, tested and true ‘free advertising’ culture. More people are listening to your music – even if it’s low quality. That means the number of potential customers goes up – or a larger fan base more likely to go to the shows where there’s high quality music being played anyway – countless people see that touring is a huge income earner for artists to begin with.

So has the transcoding problem worsened thanks to streaming? No doubt. The bigger question is, is it at the detriment of artists and listeners?

Drew Wilson on Twitter: @icecube85 and Google+.

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Scroll to Top