October kicks off the last quarter of the RNCM’s National Lottery-supported Throwing Open The Concert Doors project. Progress is ahead of schedule but there are always lots of loose ends to tie up. This feels like a good time for a quick reflection and summary of the collection’s many formats, and how we’ve approached digitising them.
The RNCM’s archive contains many thousands of concerts and masterclasses which have been superbly recorded since 1973. Back then the format was reel-to-reel; now the college records it concerts digitally. Along the way the concerts have found their way onto a variety of media, each of which need to be treated differently.
Reel-to-reel tape
These are 1/4″ magnetic tapes which play back on big clunky reel-to-reel players. They were developed after the Second World War and perfected by Swiss and Japanese manufacturers. The audio quality of the recordings can be extremely good, with a wide frequency range and very little hiss. Often even born-digital musicians prefer to record onto tape for the warm analogue sound it gives the recordings.
We play these tapes back on vintage machines into an audio converter which samples the analogue audio 96,000 times a second (double CD quality) with a bit depth of 24 (this is the amplitude or loudness granularity, which is finer than that of a CD). This results in big fat wav files of uncompressed audio which are digitally preserved, and from which listening mp3 copies are made.
In most cases these tapes will outlive us – but the urgency is the obsolescence of equipment. The machines to play them are no longer manufactured and the people who know how to maintain them are few and far between.
Some tapes from the 1970s to the 1990s were made with chemical ingredients that made them more liable to absorb moisture over time, making them stick as they pass through the machine, leaving a residue on the tape heads that makes the signal very muffled. The effect is known as ‘sticky shed syndrome’ and the only remedy is to bake them to dry them out.
However it’s a risky business which ultimately damages the tape. If you don’t get a good digital copy while the tape is dry, it will revert to its sticky condition, and actually sound much worse afterwards. For this reason it’s a balance between time and quality: is the pre-baked condition bad enough to warrant baking? How short a bake can you get away with? How much time does the project have to intervene? Does the tape still sound weird even if it’s played hot out of the oven like a pizza? That’s when you might need to rig up an IV drip to feed rubbing alcohol onto the tape as it approaches the playback head.
We’ve been unlucky to hit such a huge sticky patch but very fortunate to have a supportive funder who allowed us to use contingency funds on the equipment and time needed to deal with the problem.
More recently a new ssue has popped up: white powder syndrome. I’m not sure yet whether this is tape residue or old mould, but the treatment is the same: mask on, hoover to hand, and hope for the best! You can follow our reel-to-reel workflow online.
DAT (Digital Audio) tape
DAT tapes were one of the first truly digital formats when Sony brough them out in 1987. They are tiny, smaller than a compact cassette, yet they can store up to 2 hours of digital audio on magnetic tape. Again the tape machines are the tricky bit – they are becoming rare and the laser mechanism is very fragile. Technically the conversion is actually much simpler because the signal is already in digital form – all we need to do is match the sample rate (usually either CD quality 44.1kHz or movie quality 48kHz) and save the digital audio.
In the 1990s the RNCM noticed that the reel-to-reel tapes were going sticky and thankfully intervened to transfer hundreds of them onto DAT or CD. It’s just as well they did because the tapes would only have degraded further without intervention. The technicians left the test tones (1 and 10kHz) in at the start of the copies so we can see that the transfer is a good representation of the original tape. Then the chucked the copied reel-to-reel tapes. The college started to use DAT tapes for recording concerts rather than reel-to-reel around the same time. You can follow our DAT workflow online.
Compact Disc (Recordable)
CD-Rs became available a few years after DAT tapes and published CDs in the 1990s. They transformed digital storage and have created a massive headache for archivists ever since, because they tend to just stop working for no very good reason (some within 3 years), much sooner than their mass-produced published music cousins. They are also very sensitive to light. A laser sits underneath the disc and the signal is read in-and-out, rather than round-and-round like a record player.
The RNCM transferred some sticky 1980s tapes onto CD and also started to record to CD alongside DAT in the 1990s. We play CD-Rs on high-quality DVD-RW drives using a programme called dBPoweramp which reads the CD-Rs twice – if the copies are the same, it saves a good copy. If there are any discrepancies, the software tries again until it is happy, or reverts to ‘insecure’ copies where it interpolates sections that can’t be read. Luckily the RNCM’s CD-Rs are mostly in good condition so most of the transfers are secure.
Weirdly most of the insecure tracks are Track 1 on the CD, which is often just a test tone – nice to have for the technician but not absolutely critical for the listener. Insecure passes are fine for these tracks, but not for music. I have no idea why track 1 might be more at risk than other sections of the CD-R. Also, if a CD-R doesn’t play on one drive, it’s always worth trying it on at least one other drive. It’s a digital mystery. You can follow our CD workflow online.
Compact cassette
Philips came up with these lovely things in 1964 and they completely transformed how the world listened to music. They’re great, and much more portable than LPs or reel-to-reels, until they break or crumple and have to be spliced. My advice in this case is to try to stop breathing (and certainly no sneezing or laughing!) or else the tape will end up all over the floor.
The RNCM created these at the same time as DAT tapes, as an access copy which students could listen to on the cassette decks in the library. I have digitised hundreds of these and it’s very rare to see any indication of Noise Reduction (Dolby/NR) settings. Normally you just have to do it by your ears: does it sound better or worse with the Dolby B button switched on? It will always sound less hissy, whether it was recorded with NR or not. But you’ll lose high frequencies like ‘S’ sounds, or applause, if you play a non-NR tape through a cassette deck with NR switched on.
The other unusual thing about the RNCM’s cassettes is that they are studio quality tapes made with cobalt. These metal tapes have a much higher frequency range than the rubbish we made mix tapes with in the 1990s.
Mostly we haven’t bothered digitising these cassette copies as they just duplicate the concerts on the DAT tapes. But sometimes, if the DAT tape is missing or corrupt, they are all we have. You can follow our cassette workflow online.
So that’s my Friday brain dump of the various issues we’ve faced for each format. You can find out more about the Throwing Open The Concert Doors project at the RNCM Archive website. You can listen to lots of these recordings on Soundcloud.