Refresh on the scanner question as I am now looking for a refresh on my scansnap s1500m.
Need:
Simple document scan with pdf and need included OCR
One button scan like my current one would be nice
Be able to scan color photos
Not end of life soon.
Any experience with this? I last bought mine in 2009
I bought another ScanSnap…not the super portable one but the desktop model.
It can do color photos? Do they include OCR with the pdf scanning?
Yes, they do. OCR powered by ABBYY FineReader of some description. With mixed results.
Sad to say Adobe Acrobat is still unbeatable for OCR, IMHO.
I own a ScanSnap too, but for scanning casual photos (ie not professional) fed by a document feeder the Epson FastFoto wins my vote. It’s startlingly fast, does a decent 600dpi, can output JPEG or PDF, can scan double sided, the software offers a basic if useful approach to naming. A little pricey but recommended.
I replaced my S1500M with an ix1300. Don’t do near as many scans these days since I get more stuff electronically.
Have you used the ScanSnap iX1300 for photos? I use my S1500m for documents and love it, but it is not great for photos and is not upgradable to the new software (as far as I can tell). Plan to use it until it dies, but those things tend to happen with OS updates.
By the way, anyone using Epson FastFoto FF-680W or Canon ImageFORMULA R40?
I bought the previous model FF-640 back in 2016, and it remains terrific for high-speed photo scanning. For more detailed work, a traditional flatbed scanner can give better results, but at the cost of speed.
The only caveat with the FF-640 is that if you do a lot of scanning, you should clean the rollers and the scanning surface from time to time to avoid dust artifacts, especially if you also use it as a regular document scanner.
Yes, as mentioned above, I use the Epson FF-680W. It works just as @josehill said. Fast, terrific for certain use cases . For me I am using it to steadily work through the boxes of film prints from previous decades. These are filled with prints from pharmacies (mainly) so not a high-quality input, but a very good outcome and perfect for sharing memories of that time with family. My Epson Perfection V850 I use for scanning negatives, that’s a whole other deal.
To be honest, these days I mainly use Scanner Pro on my iPhone for document scanning. I would have to be at my desk and looking at a lot of documents to scan to bother firing up the ScanSnap.
One other method of scanning 35mm negatives I’ve been using is my Fuji X-T5 with a 30mm macro lens in combination with a Kaiser lightbox. It does a great, quick job and the Fuji’s 40Mp is very satisfying. For anything larger I use the V850.
I haven’t scanned any photos with the 1300 but did a bunch with the 1500. Quality was acceptable. Took one of the scans and printed it on a photo printer. Compared them side by side and the scanned print was just a tiny bit washed out. Still fine for snapshot type photos which I was scanning.
Sadly, it looks like ScanSnap Manager will finally bite the dust with macOS 28, what with Rosetta going away. So I have a few years before. I have to decide what software to use on my S1500M.
Who knows, maybe somebody with the help of AI will be able to port ScanSnap Manager to Apple silicon.
I was happy using like ScanSnap Manager but migrated to ScanSnap Home. Fiddling with finding the scanner on WiFi stopped with the most recent updates.
ScanSnap Manager software only will stop working in 28. No one says you have to install macOS 28. If the software runs and the scanner does its thing, fine! But if there is an aspect of 28 that makes your want to install it…
…There are other softwares that can run the Fujitsu line, I am using VueScan with an s1300i in Tahoe. It’s maybe not as simple and lovely on the screen, but I read the extensive user instructions and set it up to operate very similarly and, well, I plug in the scanner, launch the software, put the paper in the chute and press the blue button. Scans appear in pdf in the designated folder.
There was another topic where more software options for ScanSnaps were discussed… I did a search here for “scansnap” and a bunch of Topics appeared…
Thanks David. Yes, I’m aware of all that. I never install the latest OS, so I’ve got a few years remaining with ScanSnap Manager. I saw another TB article about ExactScan, which sounds like a promising candidate when a replacement is finally necessary.
Who knows, maybe somebody will rewrite ScanSnap Manager with the help of AI so that it can survive beyond Rosetta’s demise. I’ve been using it since about 2010, and old habits are hard to break.
I was shocked today (looking up how old my scanner is) to discover you can still buy new S1500Ms on Amazon!
Another big fan of the Epson FF-680W here. It has made the process of digitising 40 years of prints actually do-able. The quality is not as good as a flatbed scanner but it is good enough for the vast majority of my pics.
Before buying it I tried batch scanning prints with my Fuji IX500 but the quality was not good enough, so I bought the FF680W and used it alongside my IX500 for a while. The FF680W is a reasonable document scanner as well so I gave the IX500 to my son-in-law.
My big concern is that the Fastfoto software which is an important part of the process is not Silicon native so I keep getting the warning. I hope Epson step up.
Apple’s built-in scanning software is pretty capable.
Might be worth a look before buying anything else. I’ve only ever used VueScan with scanners that can’t be seen by macOS scanners app.
I’m interested if you’re still willing to post instructions.
Thanks,
David
Everyone here is solving how to get paper into the Mac, which is the hard half. But there’s a second half nobody’s mentioned yet, and if you’re digitising 40 years of prints it’s going to bite you: what the files are called afterwards. A folder of Scan_0001.pdf through Scan_2000.pdf is technically a digital archive and practically a filing cabinet with the labels torn off.
I should disclose up front that I make a Mac app in this space, so I’m not a neutral party. I’m not here to sell you anything — the app isn’t the interesting part of this post. This is.
While building the part that decides which way up a scanned page goes, I measured something that surprised me and that I think matters for anyone scanning in bulk: Apple’s Vision text recognition reads upside-down text almost as happily as right-side-up text. Feed it a page rotated 180° and it returns roughly the same number of characters, at similar confidence. The characters are nonsense — “JE6Ed E |EIOI” where the page says “Total a pagar” — but if your software is deciding orientation by “which rotation produced the most text”, it’s deciding by coin flip. Mine was. That’s why some batches come out with a few pages mysteriously upside down.
What works instead is asking whether the recognised words are actual words. macOS has a spell checker built in and it’s multilingual, so you can score each of the four rotations by how many real dictionary words it produced. Right-side-up scores high, upside-down scores near zero, and the difference is unambiguous.
There’s a second trap underneath that one. My first version of this scored correctly on my test pages and still failed on a user’s real letters, because my test pages had headline-sized type and his had 11-point body text. The orientation check runs on a downscaled copy, and at that size body text is about ten pixels tall — unreadable, so every rotation scores zero and the software correctly gives up. Test with the smallest type you’ll actually scan, not the type that looks good in a screenshot.
The honest limit, also measured: on crumpled or handwritten paper the text reads about equally badly in every rotation, and no amount of cleverness fixes that. On those, the software should stop guessing and let you rotate it yourself.
If it’s useful to anyone: the app is Tidy, at tidymacapp.com, $9 once. But mostly I wanted to leave the orientation finding here, because it cost me two days and it applies to any scanning workflow, not just mine.
Thanks, @sdiazthomas — Just a note for others: Sebastian has been generous in sharing advice with me about quirks with Apple’s OCR for my race timing app, where I’m planning to add some OCR capabilities to read bib numbers.
A post was merged into an existing topic: Decades After HyperCard, AI Is Democratizing Development Again