There are basically two types of audio signal. Digital and Analog. Analog can exist as purely Analog, meaning at no point does it have to come in contact with anything digital. Digital, on the other hand, has to go to analog to be heard by human ears as we have not yet invented the direct neural audio-computer interface. When we talk about the quality of digital audio it will always be dependant on the quality of the analog output.
Even the highest resolution audio is meaningless if you’re listening to it on laptop speakers. Generally, the better the analog final output the better the quality (though this might have diminishing returns but that’s an argument for the audiophile community). I’m far from being an audiophile. I love listening to music in my mastering studio on $25,000 speakers, at home on my $150 1975 dynacos and both on my $300 B&W and $20 Apple headphones.
Analog music these days pretty much solely exists as vinyl records. Tape is still used professionally though perhaps less so every year due to tape and maintenance costs. Some indie rockers are still putting out cassette tapes but good luck finding a used machine that plays it at the right and frankly I never thought cassette tapes sounded that good.
Vinyl is not only enjoying a resurgence but sales actually seem to be climbing. It’s still a drop in the bucket of CD and digital sales but almost every popular band puts out a record these days. I wonder if you’d be surprised to know that very few records produced these days have much to do with analog at all. In fact, many records are just vinyl copies of the CD. Any of the problems associated with the reproduction of CD audio would be transferred to vinyl.
Many records offer no sonic advantage other than cosmetic over the CD version. This of course is subjective but the playback of a record might just be nicer by nature but it’s essentially just a vinyl copy of the CD. In this case, instead of your digital converters and speakers being the analog delivery source, the record and player is. There are some exceptions. For example, the last Foo Fighters album was made completely analog. It was mixed down to tape and then the tape was used in the vinyl mastering. They even took it one step further and the MP3s included with the vinyl version are actually made from a recording the record. I suspect the reason for that was more to the fact that many Foo Fighters fans don’t have records players and they still wanted to sell records to them.
Other than some niche recordings the days of actual analog records are pretty much over. But, all is not lost. There is something I and other mastering engineers are doing that artists can request.
Mastering is usually done at 24bits. Sample-rates vary but few mastering engineers master at a bit-rate lower than 24. Technically, if 24-bit digital masters were provided for lacquering (the process of actually cutting the record master) the record could potentially sound better than the CD. The quality of the record could be even more improved by higher sample-rate and elimination of peak-limiting (which allows digital to be louder than analog) which is counter productive to records mastering anyway. Many vinyl records are already being mastered this way but we have no way of knowing which ones. This information isn’t provided as part of the marketing. I’m not sure if anyone would care if it was. Audio quality is not a commodity, it’s unfortunately not a selling point.
There are so many factors which play a part in the audio quality of a digital recording. Digital audio playback is only as good as it’s weakest link in the chain. If you have an amazing system but are playing back an MP3 that system won’t have the ability to work at it’s peak performance. In the same respect, if you have a high-resolution audio file and playing it back through $5 headphones that you bought on the street, it won’t matter what kind of audio file you are playing, it’s always going to sound terrible. [side-note, I once bought a pair of these thinking 'how bad can they really be' and I was actually surprised how impossibly unlistenable they were despite having no expectations of quality].
Encoding is the first element in digital playback. I talked about this in my last post so I won’t go into too much detail. Basically you have these digital encoding formats: PCM, which is the linear recording of uncompressed audio in various bit and sample-rates and DSD, which is a high-resolution only niche format format developed by Sony and Phillips. PCM can also be converted into lossless (no loss of data) and lossy delivery (MP3, AAC) formats. You need a DAC (digital to analog converter) in order to hear the encoding so the quality of the encoding is always going to be dependant on the quality of the converter.
Conversion. In order to capture analog audio recording and listen to the digital audio one must employ the use of an analog-to-digital (ADC) and digital-to-analog (DAC) converter. These converters can range from sub-$1 components in consumer audio equipment up to $20,000+ for some audiophile specially products. In fact, one the brand of converters that I use is a company in the UK called dCS and they make a CD/Digital playback system that looks like this: Vivaldi DAC ($34,999), Vivaldi Transport ($39,999), Vivaldi Master Clock ($13,499), and the Vivaldi Upsampler ($19,999). That’s over $108,000 for what is basically an amped up CD player.
Without digital conversion there is no way we can manipulate or hear digital audio. For
the most part, the better the converter the better it will sound. That doesn’t mean that cheap converters can’t sound good, some do but generally you’ll find decent converters in a recording studio and excellent converters in a mastering studio. One of the main features in Neil Young’s new Pono digital audio player is an upgraded DAC touted to be better than the iPod or iPhone’s converter.
Dither is something you may not know much about. It’s important in a discussion about digital audio because it’s used extensively in the mastering process. Most of the MP3 and CD audio you listen to that’s been produced in the last 20 years has been dithered from 24bit to 16bit. Dither, extremely simplified is adding some noise to the audio in order to avoid harsh sounds in going from a higher bit-rate to a lower bit-rate. It’s the difference between falling off a cliff and rolling down a hill. Both are no-doubt painful.
There is some debate in audio circles on wether dither needs to be used at all. I’ve listened to audio with and without dither and I tend to always prefer the dithered version but when choosing between dithers, that is the type of noise that’s used, it just becomes a whole lot of subjective opinions and no real evidence that one is better than the other. The funny thing about dither is that when you raise the level of it to where you can hear it, it sounds a lot like analog noise that you would find on a tape or vinyl record.
The best thing about dither is that you, the listener, don’t really have to think about it. Once it’s used in the mastering process (or not used) you can’t manipulate or change it.
Analog or Digital? With digital, the better the recording, mixing, mastering, encoding, playback system and analog components the better it’s going to sound. Terrible recordings won’t sound nice though good systems, great recordings won’t sound sweet with bad encoding and high-resolution encoding won’t sound any better with very cheap speakers or bad playback components. The same could be said for analog but there is something nice about listening to an old record though an old portable record player. There is nothing nice about listening to an old CD through an antique CD player with bad speakers. There is not much value in the way of digital nostalgia.