Questions, answered.

A fast, native PDF and image app for macOS and Windows. Open even very large drawings, mark them up with the full annotation toolset, measure to scale, OCR scans into searchable text, crop, organize pages, and print exactly the way you want. It also opens photographs and vector art — a site photo or an SVG detail gets the same tools as a drawing. It's built for people who work with big technical documents — plans, drawings, submittals — not just reading an e-book.

It's free while RyDF is in alpha. It won't be free forever, but there's no charge during the alpha — grab it and put it to work.

On a Mac, Apple Silicon (M1 or later) running macOS 11 Big Sur or later. On a PC, 64-bit Windows 10 or 11. Both ship from every release, so a new version lands the same day whichever one you're on. See Downloads.

JPEG, PNG, TIFF, WebP, GIF, BMP and SVG on both platforms, with SVG opening as true vector rather than a picture of one. HEIC opens as-is on macOS, so photos straight from an iPhone need nothing extra; on Windows it needs the free HEIF Image Extensions from the Microsoft Store. The start screen lists exactly what your build accepts, so it can never offer you something it can't then open.

HEIC is the one format that asks for a hand, because Microsoft ships that decoder through the Store alone — there is no redistributable copy for RyDF's installer to carry. Install the extension once and HEIC behaves like every other picture.

No. A picture has nowhere inside it to store editable markups, so saving asks which you want: convert to a PDF and carry on there with everything still editable, or render the markups into a new picture to send someone. Both write a new file; your original is untouched either way.

Opening an SVG keeps its real geometry instead of turning it into pixels, so it stays sharp however far you zoom and its text is still selectable and searchable.

Exporting to SVG works the same way round: paths stay paths, text stays text, and your markups come across as real SVG shapes. The document's own clickable links aren't carried across. A few things vector genuinely cannot express — gradients, blend modes, soft masks. Where RyDF meets one it renders just that part and places it as an image in exactly the right spot, then tells you it did. It won't quietly redraw a gradient as a flat colour.

No. Cropping moves the page's boundary rather than throwing pixels away — the rest of the sheet is still in the file, and ⌘Z takes it back. That matters most on a photograph, where a crop that discarded the edges would be something you couldn't undo once you'd saved.

It marks the parts of the picture that are transparent, the same convention other image apps use. Without it, white artwork on a transparent background would be white-on-white and invisible. It's only ever shown on screen — it isn't part of the picture, and it never prints or exports. Change the colour, pattern and check size under Preferences ▸ Transparent images.

Windows, yes. macOS (Apple Silicon) and Windows (64-bit, Windows 10 or 11) both ship from every release, built from the same code — take either one from Downloads.

Linux isn't shipped. The core engine is cross-platform, so it isn't ruled out, but nothing is promised and there's no date to give you.

The Windows build is the younger of the two and has a few rough edges the Mac build doesn't: the installer isn't code-signed yet, HEIC needs a free codec from the Microsoft Store, text recognition needs a language installed, and two print options aren't passed to the driver. Each one is answered on this page.

The part that matters most does. Both platforms send real vector output at the driver's own resolution, so linework prints as linework rather than as a picture of itself, and page range, copies, orientation and scale all work on either one.

One thing isn't passed through on Windows yet: the paper. The driver chooses it and nothing RyDF sends can change that, so the Paper size control is switched off there and the job is laid out for whatever sheet the driver reports — set the paper in the printer's Windows settings and the print follows it. RyDF doesn't read the driver's duplex list either, so Two-sided isn't offered. Copies always come out collated — Windows prints each copy as a whole document, so unticking Collate there makes no difference — and greyscale works in Raster mode. The print queue beside the Print button is macOS-only too: it reads CUPS, which Windows doesn't have, so the queue button never appears there and a job that has already been spooled has to be stopped from Windows' own print queue.

On macOS the job goes through CUPS with the whole set passed through: paper size, collate, greyscale, reverse order, duplex, page range, copies, orientation and scale.

Yes. RyDF takes Ctrl for everything the Mac takes ⌘ for, so the shortcuts written on this site read straight across — ⌘S to save is CtrlS on Windows, and so on down the list.

Into the PDF itself. When you press ⌘S, RyDF writes your annotations back into the same file — there's no separate sidecar to keep track of or lose. Reopen the PDF in RyDF later and every markup is still fully editable.

A picture is different, because a JPEG or PNG has nowhere inside it to keep editable markups. ⌘S on one asks whether to convert to a PDF — carrying on there, with the markups still editable — or to render them into a new picture. Either way a new file is written and your original is left alone.

Yes. A plain ⌘S writes both — real PDF annotations that Preview, Acrobat and Bluebeam display, including lines and arrows, and RyDF's own exact record tucked inside, so the file stays fully editable when it comes back to you. Use Export ▸ Save a Copy… when you want to leave the original untouched and send a separate file. See the user guide for which save to use when.

No. Nothing is written until you save. If you'd rather keep the original untouched, use Save a Copy and mark up the copy. Between saves, a lightweight autosave keeps a backup so you don't lose work.

Opening a picture never modifies it: both save options write a new file. RyDF also refuses outright to save a document over a file that isn't a PDF, so a mistyped filename can't overwrite a photograph with PDF data.

That's the point. RyDF renders pages in tiles and keeps a RAM cache of what you've viewed, so huge multi-hundred-page drawings scroll and zoom smoothly. Files open straight to the first page while the rest loads in the background.

Yes — each opens in its own tab, and switching between them is instant because they stay loaded in the background.

That holds however you open them — from inside RyDF, or by double-clicking in Finder or Explorer, one at a time or several at once.

As accurate as your scale, and there are three ways to set one: draw a line over a known length and type what it measures, type an architectural ratio straight in (Set scale from ratio — 1/4"=1', no line needed), or let RyDF read the scale a sheet states about itself, which it does automatically where the drawing records one. Your own setting always wins. Snap-to-geometry helps you land exactly on the drawing's own linework, and scale is stored per page so a set with mixed scales stays correct.

Measurements display in millimeters, inches or points. When you set a page's scale you can type the unit you mean — mm, cm, m, in, ft, pt, or " and ' — along with simple math and fractions like 1/8" or 3 1/2'.

No. OCR runs entirely on your own machine — Apple's Vision framework on macOS, Windows' own recognizer on Windows. RyDF adds an invisible text layer over the scanned image, so the page looks identical but the text is now selectable and searchable. Your document never leaves your computer.

There is nothing to install on macOS. On Windows it needs a recognition language present under Settings → Time & language → Language & region.

On a drawing it reads each page three ways round, so vertical title-block labels and dimension strings are caught as well as horizontal text — that's Most accurate, the mode it starts on; Fast makes one upright pass, takes about a quarter of the time, and is right for ordinary prose. Large sheets are read in overlapping tiles, which is what makes 4–5 pt dimension text legible to the recognizer at all.

It shouldn't — RyDF opens normally. If you somehow hit it on an old build, right-click (or Control-click) RyDF → Open once, or just re-download the latest from Downloads.

No — expect it on the first run. The Windows installer isn't code-signed yet, and SmartScreen warns about any installer it can't tie to a publisher it knows. Choose More info, then Run anyway.

Signing is on the list. Until it's done, take the installer from Downloads or from RyDF's own updater, so you always know where the file came from.

RyDF checks GitHub Releases for newer versions and downloads them for you — on a schedule you pick in Preferences, or only when you ask, with Check for Updates Now. It fetches the right installer for the platform you're on and leaves it in your Downloads folder. The updater works the same way on macOS and Windows.

On Windows there's nothing to do: RyDF runs the installer, which shows a progress bar and no questions, and the app reopens itself on the new version. On macOS it opens the .dmg and quits, and you drag RyDF to Applications over the old copy. You can always grab a build by hand from Downloads instead.

Yes, and it asks once when you first launch it: Yes, No, or Ask Me Later. No is permanent and won't be raised again; Ask Me Later asks again next launch. You can settle it whenever you like under Preferences ▸ Default PDF viewer.

On macOS, Yes sets it there and then. On Windows it can't: since Windows 8 no application is allowed to claim a file type for itself, so Yes opens the Windows Default apps settings page and you make the change there. That's Windows' rule, not a RyDF bug.

RyDF is a local app. Your PDFs stay on your own machine — it doesn't upload them, and OCR is on-device on macOS and Windows alike. The only network calls it makes are to GitHub to check for updates.

Perfect — that's exactly what an alpha needs. Use the feedback form and tell us what you were doing and what happened (a screenshot helps a lot). Feature ideas are welcome there too.