Ten years or so ago, I worked on a project to digitise tapes for a large archiving company whose record company client had suffered a major fire in one of their own internal archives (nothing to do with the company for which I was working) and lost a lot of tapes. They weren't sure what exactly ahd been lost so spent over a year retrieving every tape they could find which might be pertinent from all over the world - territorial production masters from production plants, media playout copies, listening copies held in archives/museums, private/artist/record company collections, anything which might enable them to recover otherwise lost early generation copies of their masters.
The copies ranged from acetate, and even paper, tapes to 1/4" and 1/2" analogue reels, DAT and PCM1600/10/30 U-Matic masters, Sony DASH and Mitsubishi PD format open reel digital, even some PCM-F1format tapes.
The digitsation was split between a few UK based companies and the one which hired me built, in a matter of weeks, thirteen networked transfer studios and all the necessary support facilites - dedicated maintenance workshop, high tech fire-safe environmentally controlled media and secure data storage vaults, multi station scanning facility, process management and admin backup, etc., etc. - specifically for the project, then employed enough people to run it 24hrs a day.
Everything was catalogued and digitsed - 24/96 copies of the analogue audio, whatever was appropriate for the digital audio, and high res scans of the media boxes and any associated paperwork - with the final output of the project ending up on some large HD storage arrays and multiple LTO tape backups.
When it was all done, the facility was closed, dismantled and the equipment mostly returned to the parent company in the USA with just a few studios being rebuilt in another company building in the UK.
We transferred a lot of audio, with the earliest stuff dating from the 1940s, going up to things I'd worked on in the 1990s and ranging from straight playbacks which just worked, to things requiring varying levels of physical restoration and recovery of media before they could go anywhere near a playback machine. All good fun and really interesting. Not least for the mix of engineers working on the project. On several occasions I'd give a tape to an engineer and they'd say oh, I made this. One fun moment was when I was trying to decipher the handwirting on a tape box from the Apple Corps studio, recorded on the day I was born. I showed it to the engineer standing next to me and said 'how is anyone supposed to read that' only for him to reply with all the info. I asked how he managed to read it and he grinned and said 'Easy - it was written by an 18 year old me. I was a tape op there". Fun project.
And on digital transfers, I'm currently (re-)cataloguing an archive, about 99.5% of which I recorded for a broadcast client so they can see what still exists and see whether they might want to re-broadcast any of it. The several thousand DATs and CD-Rs date from 1992 onwards but unfortunately the database which held details of the content has been lost/corrupted to the point where it's of very limited use. So far, the DATs are mostly transferring fine - failure rate of 2 tapes out of over 400; sadly the same can't be said of the CD-Rs which are showing problematically high error rates and various drop-outs in at least 50% of cases. The results will probably vary a lot as I work on more differently aged media. Most of the DATs so far are from 1995 and the CD-Rs from around 1998-2001. (The DATs have all been BASF and the CD-Rs various types of Taiyo Yuden and everything has been stored in the dark in a stable environment.) Thankfully there have been backups and multiple copies to provide cover so far. Glad I kept everything now!