Saturday 3 October 2015

An Idiot's Guide To Becoming Half An Audiophile

If you have been here over the course of the few months, you would have noticed me harping on dynamic range; or the loudness / compression on songs. I realised that not many people understood what I was getting at on production, so I decided to do a post on production itself and I hope this helps you understand the situation.

Note: For the post, I will not be showing or using spectograms simply because I don't have the app Audacity on my laptop and am honestly too busy with other stuff to learn this and post on it. Perhaps in the future.

Loudness War- Why you should be concerned.

There it is, the loudness war. For beginners, the loudness war is a phenomenon where music producers start to over-compress music in the recording studio itself. This phenomenon was pioneered by once revered producer Rick Rubin. The bearded vegetarian behind Slayer's earlier material started overcompressing (the correct term is 'brickwall') music so that it sounds louder and more impactful to the average listener with average earphones and average listening skills. The measurement unit for this compression is dynamic range, or abbreviated as 'DR'. DR is measured on a scale of 1 to 20, where DR1 is the most compressed and DR20 is the least compressed. For example, a quiet, ambient wind sound would be DR20, and putting a metal spoon in a blender and turning it on would probably be DR4 (which is actually less compressed than some of the songs I have reviewed on this blog). Why should you be concerned? For starters, overcompressed music is bad for your ears. Do you ever hear (or feel) a sharp static noise whenever you increase the volume on your song? That is a result of poor audio practice, not just on your part, but the musicians and producers. This is the reason you cannot enjoy your music loud, and that IS a problem for certain genres, chiefly, metal. When using cheap earphones with low power ratings, your earphones tend to lack the impact that a hard hitting song should have- which is why producers have beefed up compression on their end, so that you would get that 'bass' effect on your earphones. This would have been around the 90s, where the CD started to overtake vinyls and cassettes. I will explain the compression on CDs in a while,but the gist is this: CDs have more compression than vinyls, obviously because of the differences in size of either medium. This has snowballed towards the 2000s and its effects are being seen today. Ask your parents, dig up any older songs of bands from the 70s and 80s, you will hear a very big difference in the sound. Older music is often considered 'warmer', which is the layman way of acknowledging its higher dynamic range. Don't believe me? Which of the two songs below is louder?








Of course, if you picked Justin Bieber because you know I am trying to prove a point, then you are right. Metallica's Master Of Puppets is DR13, while Bieber's Baby is DR6. That's right guys, Justin fucking Bieber is louder than the most famous heavy metal song.

This is why you should be concerned, the difference between brickwalled music and dynamic music is huge. In Metallica's MoP, you can hear the separate instruments, you can hear/feel the reverb from the bass and the drums, and it actually sounds monstrous on a proper sound system. On the other hand, Justin Bieber is hollow, cold and too dull on a pair of good earphones, though it sounds 'loud'. Especially so in heavy metal, where instrumental prowess is a prized asset to musicians, you need to hear every separate instrument. This also applies to jazz and classical music, or any genre that actually requires some intelligence.

Still not impressed? I have uploaded some high dynamic ranged songs on my Dropbox folder here, and if you have a decent pair of speakers or earphones, you can actually listen and see for yourself. Bass guitar is audible, drums sound tighter and more powerful when all the reverb is brought back and the song sounds pleasant no matter how loud you play the music. It doesn't hurt your ears.

Note: The loudness inherent in the song (compression) is not the same as the loudness on your speakers or headphones (volume). Playing the song loudly is not the same as playing a loud song.

The Details:

You will need to understand some jargon here before I move on.

DR - Dynamic range. As mentioned earlier, this is the measure of how compressed a song is.

Bitrate - This one is tricky because people who listen to MP3 pretend this decides how great the quality is. The bitrate is the amount of bits, or information, per second on the song. In other words, the level of 'detail', so to speak. The problem is that a higher bitrate doesn't necessarily entail a better sounding song, something I will show later. In any case, this term isn't as important as I thought it was.

Sample Rate - This is simply the amount of frequency you pack into the song file. For most MP3 and CD formats, the sample rate is 44.1kHz. The next increment is 48kHz, which you will probably see on higher quality lossless formats. The jump from 44.1kHz to 48kHz isn't drastic, but it does result in a better dynamic range, so it does sound better. the next level is 96kHz, which is what you commonly find on vinyls and super-audio CDs. At this frequency, it's usually at a very high dynamic range- until recently. In the past, artistes used to have separate mixes for vinyls and CDs, but nowadays, people apply the same mix to save time. This wouldn't be a problem if it were mixed for vinyl and applied to CDs, but the reverse is done. In other words, most modern day vinyls are no better than your CDs, so fuck all the hipsters out there. The last level is 192kHz, this is the frequency at which the musicians record and mix in the studio. The irony is that this technology wasn't prevalent up till the 90s, which is when overcompression started to take its initial form. Higher level vinyls still contain this sample rate.

MP3 - the most common music format. It's often called lossy. Why? Recall that the musicians record at 192kHz from the above paragraph, To form mp3, the higher and lower end of the frequencies are cut off, leaving the middle 44.1kHz of frequencies. The people who do this argue that you don't need the tail ends of the recorded frequencies. Of course, in isolation, they sound negligible. But on better earphones, you realise that this takes away a lot of the natural reverb. This is why the bass sounds non-existent, the drums sound thin, the cymbals sound dull and lifeless. That's the entire rhythm section affected. For metal, this is why your blastbeats sound like a blur. That's not what they should sound like. Yes, the main song is untouched, the vocals are still there since humans can only sing within that frequency range, but the song itself sounds like its covered in dust.

Flac - Free Lossless Audio Codec. Remember again that audio is recorded in 192kHz? For lossless music, the sound is compressed from 192kHz to 44.1 (this is the most common ones, you can still find it in 48kHz). This is the same sample rate as on mp3, BUT nothing is lost (hence the name). This leads to the problem I explained on my introduction, though - the compression. This is also why a lot of people say there is no difference between mp3 and Flac, since the only difference is the bitrate (usually around 700-1000+ on flac). I highly disagree- with better audio codecs (I use an iPod), you can actually hear a pretty big difference between the two formats. Some of the biggest jumps in audio quality occur in black metal and progressive rock, at least in personal experience. Again, with  badly produced songs and bad earphones, the average listener is not going to be able to pick up the significant changes in sound.

Recently, I have started to compare flac derived from vinyl with flac derived from CDs. There are devices that 'read' your vinyls and convert them to digital information at 192kHz, so nothing is lost. For this post, I decided to compare with Cannibal Corpse's album, Kill.

How the CD files look on my iTunes

How the vinyl files look on my iTunes



As you can see, despite having the same sample rate, the vinyl rips are FAR more dynamic in the dynamic range. I usually keep my files at 44.1kHz, because one album at 192kHz can be around 2GB, and I don't have the luxury of that much space on my laptop. Regardless, the vinyl derived music is a lot clearer, despite sounding softer. The songs sound warm and clear despite... well, despite it being death metal. That isn't to say it sounds a lot less impactful- the reverse is true. The bass drums pound harder than ever and you can actually hear the tightness of the strings, be it guitars or bass, in the music. In other words, it sounds like the music has been dusted; thicker bass, tighter drums and piercing symbols and guitar solos. The payoffs to good audio practices are tremendous.

I might talk about the process of converting these files, and my own audio equipment in another post.

No comments:

Post a Comment