Simon Bradshaw (
major_clanger) wrote2011-01-31 01:14 pm
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
Entry tags:
Slide Scanning
I've got a pile of old 35mm slides, including some nice Kodachromes from a school trip to Pompeii back in 1982. Does anyone have experience of slide-scanning services and any recommendations?
no subject
no subject
no subject
no subject
no subject
no subject
50 quid buys you a usable scanner based on phone camera imaging and screen, dumping to an SD card (try Farnell). Not great, but quick. Two six-slot push-through magazines, so you can re-load one whilst processing the next.
Next step up is my £300 HP slide scanner. Nice bit of kit, even nicer if you strip the mounts, but it still needs a Windows 95 machine for its pseudo-SCSI drivers (I keep one for just this).
Paper scanners make very poor slide scanners, even if they did come with an add-on light box.
The ultimate is someone like Redcliff in Bristol (who we used commercially). They strip the mounts, wash, and use an oil immersion scanner. Best quality, but $$$$! Commercial services who strip the mounts cost real money, scanning whilst in the mount is no better than DIY grade.
IMHO, best approach is two-pass scanning. Scan the lot quickly with a cheap "staring" scanner, then re-do the dusty ones and the good ones, ideally with a better quality scanner. I don't wash mine before scanning, but I do wash and dry whole boxes if they turn out to be dirty (a dirty slide is usually a boxful of dirty slides). I also strip mounts (and scan as bare film) if they're old card mounts, or Agfa that have cockled. I usually strip glass mounts too, because my glassed slides were the good stuff, but they're also well-travelled and probably dirty. Some scanners can be a problem with glassed slides, even if they're not showing obvious Newton's rings.
You also need organisation and a plan for metadata. Minimum metadata is an auto-allocated unique ID for each box and the ID within it (just 1-36 is OK). I catch manual metadata at the box level (from old sticky labels on the box ends, date / location / filmstock etc.) and only rarely add more to this per image (titling the good / interesting stuff, recording anything previously sold / competition winning). Some crude Python (or Perl) shoves this as a batch into the EXIF for each image. I also dump XML versions per directory (box) and can (manyana) populate this into a central database.
My metadata use Dublin Core / RDF / Bath profile standards, so it's easy to move it around later. The box IDs are pre-printed on some new sticky labels, which also identify boxes as "scanned".
Teenagers are useless for outsourcing, elderly aunts are far better.
no subject