The following MBW blog comes from Kieron Faller (pictured). Kieron is GM of CI, a London-based digital music delivery company that was founded in 2003. Now delivering to Apple Music, Spotify, Google, Amazon and dozens of other digital music platforms, CI are a trusted partner for almost 150 clients, from huge distributors to tiny labels, across the UK, Europe, North America, Asia and Africa. The firm delivers content to over 200 download and streaming services, and other places including Shazam, Gracenote and AudioLock.
The company I work for has been around since before iTunes even existed.
CI’s founders saw that the independent sector of the music industry was going to have a tough time meeting all of the technical needs of the new digital world that was opening up in the wake of Napster.
After all, these were music companies, not technology companies. Why should they know how to make a data package and send it over the internet?
CI has seen the evolution of digital, from the days of 96kbps mp3 downloads to CD-quality streaming and hi-definition downloads.
We’ve helped shape metadata standards across the industry, and fought the indies’ corner when the major labels have tried to use their scale to tilt the technical playing field away from fair competition.
We know that the future music product will be made up of three key factors, just as it is today: (i) digital audio files; (ii) metadata; (iii) artwork.
If you are involved in creating music products, you’ve got to get all of them right to achieve true excellence in the future marketplace.
And that means ensuring that your finished master product has three key attributes:
1) Digital audio files of 24bit, 96kHz quality
If you want to know the detail of why 24bit, 192kHz audio is not worth the extra time or effort to store, transmit and play back, take a look at this excellent explanation.
The headlines are that the human ear cannot hear that level of detail (as proven in scientific double blind test conditions), and that inaudible frequencies in the sound wave might interfere with audible frequencies (due to physics) and actually degrade the audible part of the sound wave.
In the vast majority of scenarios, CD quality 16bit, 44.1kHz audio is as good as the human ear can hear.
Additional audio detail is not distinguishable from the CD quality audio. As an archival format, to remix or remaster from, a 24bit, 96kHz file will be easily good enough.
“As streaming makes its way further into the home via connected speakers, there is a case for CD-quality audio.”
More detail on one of the best blind tests can be found here.
The saying ‘good enough is good enough’ comes to mind here. Once audio quality gets to CD quality, anything ‘better’ than that in a technical sense adds no improvement to the experience of listening.
Almost all streaming services provide music at 320kbps mp3 or below right now, so there is some work to do to move up to CD quality streaming, although given that so much streaming usage involves smartphone audio chains and headphones or earphones of questionable quality, one could ask whether going above 320kbps mp3 files will actually make a qualitative difference to most listening.
However, as streaming makes its way further into the home via connected speakers, technology such as Gramofon, and Smart TVs, there is a case for the CD-quality streaming service.
It is a niche that Tidal should be making the most of, given that it is the only service currently offering CD-quality streaming.
2) Complete and perfect metadata
As for metadata, there is one cardinal rule to remember. In the digital world, metadata is how you get paid.
Payments for downloads or streaming usage, either direct from services or via a collecting society like SoundExchange or PRS, depend on traceability. On the sound recording side, barcode and ISRC are utterly vital.
If one of these becomes incorrect somewhere along the chain, funds will be going astray, either to a black box of unattributable income, or to one of your competitors, who may or may not realise the error and may or may not hand the money back.
There is a more subtle way metadata has a big impact on how you get paid too. In the digital age, a huge amount of consumption, especially on streaming services, is via the search bar.
Users type in the song name, or artist name, or album title, to find the music they are looking for.
“If you are using an intern (or team of interns) to deal with your digital metadata, stop.”
If sometimes the artist name is spelled differently from album to album, or if there is inconsistency in the searchable metadata across different music services, this can result in ‘lost listens’.
Such issues also add cost and complexity to the whole supply chain.
Do you really think each music service should have to employ a team of people just to check and correct metadata because people earlier in the supply chain have not taken enough care and attention over getting it right in the first place?
This is money off the bottom line of Spotify, Deezer, Rdio, iTunes, etc etc. Without this cost incurred, they would be able to pay rightsholders (and by extension, artists) more.
If you have an intern (or teams of interns) dealing with your digital metadata, stop.
Build a proper team who understand the importance of this issue to your business, and treat it as the vital part of the process that it is.
You wouldn’t let an intern produce the record, so why would you let them mess up its journey to the fan, and imposing costs on every part of the music industry value chain too?
3) Artwork suitable for a 4k screen
We’ve seen iTunes’ requirements for artwork size move in parallel with screen technology.
It used to be the case that 800 pixels square was enough. The iMac G4 on sale in 2003 had a screen resolution of 1,440 x 900 pixels. That meant that an image 800 pixels square would fill half of the screen by area, and would be easily big enough for inclusion in the iTunes user interface alongside play controls and product metadata.
On a 4K display (3,840 x 2,160 pixels), the same image would be just 8% of the screen area. With the advent of Smart TVs and devices such as Apple TV, Google Chromecast and Amazon Fire Stick, streaming services are going to be blasting through the best sound system in your house (which takes us back to the audio issue) and also seen on the biggest screen in the house.
This is tough for the streaming services – they have to optimise for 5-inch screens and 55-inch screens simultaneously.
In contrast to the audio issue, differences in visual quality are easily discernible to the average user.
Samsung and Apple smartphones are now 1,440 pixels across the top of the screen, so a 1,400 pixel artwork file can’t even stretch all the way from side to side of the screen without being scaled up slightly. Here, such changes may not be too noticeable, but on a tablet or TV, pixelated scaled-up images are going to give users a poor experience.
“There is a danger of major label gain here: They have the resources, processes and foresight to undertake systematic upgrading of their back catalogue artwork.”
Some services, such as Google Play Music, are attempting full-screen product artwork visualisations while the music is playing, but because the music industry has always done just what was necessary in the digital space to keep up with contemporary needs, there is little back catalogue digital artwork at more than around 1,200 pixels square.
This is a real issue for music services trying to provide an immersive, or just a great, experience on a 4k TV (pictured, main).
For reference, a 12-inch record sleeve at 300 dots per inch in the real-world printing jargon, is 3,600 dots. So the 4,000 pixel x 4,000 pixel target is not crazy.
If your artwork is being designed digitally for vinyl too, there should be at least a 3,600 pixel version in existence somewhere. You at least need to find it, keep it somewhere secure and findable later, and use it as your digital artwork too.
And in future, brief your designers that the final master artwork should be the 4,000 pixel square size.
Importantly for the independent sector, there is a danger of a major label gain here.
They have the resources, processes and foresight to undertake systematic upgrading or re-creation of their back catalogue artwork to prepare for this pixel-dense world.
If their products look better in the user experiences of these services, there is a cue regarding the quality of the product.
The user may think (perhaps subconsciously) ‘this product is better than the other one, because this one has had more care and attention lavished on it’.
We all carry out this sort of quick analysis of products every day, essentially automatically, and this kind of quick quality shortcut to judgement will hurt those labels and artists who have not taken the time and effort to create artwork that will look stunning on a big screen.Music Business Worldwide