"Tim" == Tim Keller <turbofx@gmail.com> writes:
The problem is one of organization. Let's imagine you get these all scanned in, now you've got +/- ~19,000 pictures to sort.
Same with the tapes... just cause I know which tape has which files tells me nothing about how useful that file will be... it's always a pain to archive an old ASIC chip design because you need: 1. design files 2. the specific tool versions 3. OS and hardware for the tools 4. licenses for the tools! - good luck when alot of the vendors are bought out or shutdown 5. someone who knows how to run steps 1-4. 6. the same process at the FAB running to make the chips... So archiving a design is not simple or cheap. You really don't want the intermediate files honestly, just the initial VHDL and rough layout tools are probably enough for anything older than 5 years since you'll have to re-spin the workflow anyway, unless you're lucky enough to still have it hanging around. Once you hit ten yeats... all bets are off. Now with scanning in those slides, I'd just toss them into 'photoprism' and let it classify them for me. Will it be perfect? No, but it did a decent job on my 70,000+ images I tossed at it. Took a few days since I don't have NVidia CPUs to run the classifier on for a speed up. But it did a neat job. And it even find obects as well. Very cool stuff.