Old magazines on an e-reader: archive.org scans rebuilt for Kindle Scribe, Kobo and Boox
Archive.org holds a century of magazines, newspapers and zines as scanned PDFs, and nearly all of them are the wrong shape for an e-reader: two pages to a scan, a grey cast from the microfilm or the yellowed paper, print margins, and a file size the reader chokes on. Open one on a Kindle Scribe and you get a spread shrunk to the width of the panel, with type too small to read and a page turn that takes three seconds.
↑ Upload and configure in our Studio
Radio News, July 1926, Hugo Gernsback (editor), typeset by the Internet Archive, public domain
This converter treats every scan as an image and rebuilds it for the panel you own. It splits a two-page spread into two pages, trims the margins so the column fills the screen, pushes the grey of the scan towards black on white so the type survives e-ink, upscales a low-resolution scan to the panel’s own pixel count, and writes a file the reader opens natively: EPUB for a Kindle, KEPUB for a Kobo, or a CBZ of the cleaned pages if you read in a comic app on a Boox. A 100 MB archive.org download comes out as a file sized for the device, which is why page turns are fast again.
Honest scope first: this is for scans you hold as a PDF or as page images. It does not run OCR, it does not reflow the text, and a blurry microfilm scan stays blurry. It does the part that makes a scanned page legible on e-ink, in the browser, free, with a live preview before anything converts.
- Files auto-delete after 24 hours
- No install, runs in any browser
- Free to use
- The genuine open-source KCC engine
How to optimise an archive.org PDF for e-ink
Drop in the scan
The archive.org PDF as downloaded, up to 1GB and 1000 pages per file and 20 files per batch. A ZIP of the page images from the JP2 download, or a CBZ, is accepted too, so a run of issues can go in as one batch.
Crop, split, contrast and upscale are already set
This page presets margin cropping with page-number trimming, spread splitting with a rotated copy of each spread kept for full-width illustrations, an auto-level pass that pushes the grey of the scan towards black and white, and upscaling, so a 100 dpi scan is rebuilt at the panel’s pixel count rather than stretched by the device. Manga mode is off, so pages read left to right.
Pick your device and the output
The Kindle Scribe profile is preselected as a starting point. Every current Kindle and Kobo has its own profile that builds pages at that panel’s exact pixel count, so pick your model; for a Boox or any reader not in the list, choose Custom conversion and type the panel’s width and height in pixels. EPUB is the default for a Kindle and KEPUB for a Kobo; pick CBZ if you want the cleaned page images for a comic reader app.
Check the preview, convert, download
The preview renders real pages at the device size, which is the moment to confirm the smallest body type is legible and the spreads were split where you expected. Convert, then download the file or pick it up on the reader from your Reader Shelf over Wi-Fi. Files auto-delete after 24 hours.
Which e-readers work for scanned magazines, by brand
| Device | Profile in the picker | Screen | Recommended output |
|---|---|---|---|
| Kindle Scribe | Kindle Scribe 1/2 | 10.2 inch, 1860 x 2480; a magazine page at close to print size | EPUB, delivered by Send to Kindle, the Reader Shelf or USB |
| Kobo Elipsa and Sage | Kobo Elipsa, Kobo Sage | 10.3 and 8 inch | KEPUB, copied over USB or downloaded from the Reader Shelf in the Kobo browser |
| Boox, PocketBook and other Android readers | Custom conversion | Type the panel’s pixel width and height | CBZ for a comic reader app, or EPUB or PDF, downloaded from the Reader Shelf |
| reMarkable | A profile per model, reMarkable 1, 2 and Paper Pro | 10.3 to 11.8 inch depending on the model | PDF, loaded through the reMarkable app |
| Any 6 or 7 inch reader | Your model’s profile | Readable for a single column once the spread is split and the margins are cropped, small for a dense page | The brand’s native format, as above |
What this cannot do
No OCR, no text search, no reflow, and no sharpening of a bad scan: the output is a book of page images at the panel’s size, which is exactly what makes it fast. A newspaper broadsheet scanned as one image stays a broadsheet; cropping and splitting help, but a six-column page on a 7 inch panel is small whatever you do to it.
Check the scan’s rights first
Archive.org marks most pre-1929 US periodicals as public domain and many later uploads carry a Creative Commons licence; some scans are there without a rights statement at all. This converter does not care, but your conscience might. The demo on this page is a 1926 issue carrying the Public Domain Mark.
Right to left is off here
Every manga page on this site defaults to manga mode, which stamps right-to-left reading into the output and would put the right page of a split spread first. Magazines read left to right, so this page turns manga mode off before your file goes in; you do not need to find the switch.
Frequently asked questions
Why is an archive.org PDF so slow and small on my e-reader?
Because it is a stack of full-resolution page scans, often two pages to an image, wrapped in a PDF. The reader has to decode a 5 MB image per page turn and then shrink it to fit the panel, so you get a slow turn and a tiny page. Rebuilding every page at the panel’s own pixel count fixes both: the file carries exactly the pixels the device shows, and nothing is shrunk at read time.
What happens to two-page spreads?
Each spread is split into its left and right page, in that order, so you read one page at a time at full width. The preset also keeps a rotated copy of the whole spread, so a double-page illustration or a centrefold is still there to look at sideways. If a scan is already one page per image, the splitter leaves it alone.
Will a grey microfilm scan look all right on e-ink?
Better than the raw file. The auto-level pass pushes the grey background towards white and the type towards black, which is most of what makes a 1920s scan readable on a greyscale panel, and the preview shows the real pixels before you commit. What it cannot do is add detail: a scan that is blurry at full size is blurry on the device too.
Which output should I pick: EPUB, KEPUB or CBZ?
EPUB for a Kindle, which is what Send to Kindle accepts. KEPUB for a Kobo, for faster page turns. CBZ if you read in a comic app on a Boox or an Android tablet, because it is simply the cleaned page images in a ZIP. PDF is there too, for a reMarkable or for a Boox whose reader prefers it.
Can I read the result in colour?
A colour magazine cover or a colour illustration can be kept: turn on Keep colour in the options before converting and pick a colour-panel profile such as the Kindle Colorsoft or the Kobo Libra Colour. This page’s preset leaves colour off, because a greyscale panel shows greyscale anyway and the file is a third of the size.
Page extraction, spread splitting, cropping and contrast run on Kindle Comic Converter (KCC), the open-source engine built for putting image-based pages on e-readers. A scanned magazine page is the same kind of object as a comic page, which is why the fit is honest. Hosted here with a live preview, no install.
Free, spreads split, margins cropped, built for your panel