Why the print dialog is the safest converter
A print-to-PDF export is produced by your own machine from the page that is already rendered in front of you. Nothing is uploaded, no third party sees the document, and there is no installer that bundles something you did not ask for. It is also the only conversion route where the preview you approve is genuinely the file you get, because the same layout engine produces both.
The trade-off is honest and worth stating plainly: you are exporting a rendering, not translating a file format. Text stays selectable and searchable, links usually survive, but anything that depends on the original application's pagination is being re-flowed by the browser.
Opening the dialog
Use Ctrl+P on Windows and Linux, or Cmd+P on macOS. In the destination list choose Save as PDF in Chrome or Edge, Save to PDF in Firefox, or the PDF menu at the bottom left of the macOS print sheet in Safari. On Windows, "Microsoft Print to PDF" also works, though it goes through the system printer driver and tends to give slightly different margins than the browser's own exporter.
On iOS and Android the share sheet is the shortcut: choose Print, then pinch outwards on the print preview (iOS) or pick "Save as PDF" from the printer list (Android) to get a file instead of a job.
The four settings that actually change the file
- Paper size. A4 (210×297 mm) and Letter (8.5×11 in) differ by about 18 mm of height and 6 mm of width. That is enough to push a final line onto a new page. Match the size the recipient expects rather than the default from your locale.
- Margins. "Default" is roughly 10 mm in most browsers; "None" removes the safety area that physical printers need. If the PDF may ever be printed on paper, keep at least 10 mm.
- Scale. "Fit to page width" is convenient on screen and unpredictable across viewers. A fixed 100% scale produces a file whose measurements you can reason about. Reduce to 90% only when a table genuinely will not fit.
- Background graphics. Off by default. Leave it off for text documents, because printing backgrounds is what turns a clean report into a grey-blocked one; turn it on only when a coloured table header carries meaning.
Headers and footers is a fifth option worth a moment: the browser stamps the page URL and date into the margin. That is useful for an archive copy and inappropriate for a document you are submitting.
What moves, and why
Page breaks are the usual surprise. A word processor decides pagination from its own layout model; a browser decides it from CSS and the paper box you just chose. Tables split across pages, a heading can end up orphaned at the bottom of a page, and footnote positions are not preserved at all. Fonts are the second surprise: if the original relies on a typeface that is not installed locally, the browser substitutes a metrically different one and every line length shifts.
Three things are frequently expected and do not happen. Comments and tracked changes are not exported, because they are not rendered. Form fields become flat text rather than fillable PDF fields. Passwords and permissions cannot be applied from the print dialog; that needs a dedicated PDF tool.
Making the export reproducible
If you regenerate the same document repeatedly — a weekly report, a set of invoices — write your settings down once: paper size, margin preset, scale, background on or off, headers off. Print dialogs remember the last used values per browser profile, not per document, so a colleague following the same steps on a different machine can produce a visibly different PDF unless the settings are recorded.
Zoom level in the page is a related trap. Browser zoom does not usually affect print output, since the print stylesheet re-lays out the content, but a page that positions elements from the viewport size can render differently. Reset zoom to 100% before printing anything whose layout matters.
Where Document Forge fits
The document tool on this site prepares the print view for you. It extracts the readable text from an HWPX archive, or takes plain text, Markdown, or sanitised HTML, and renders it as a clean single-column document with sensible print margins — then hands that to your browser's own print dialog. Nothing is uploaded in this path. An existing PDF is opened in the browser's built-in viewer instead, because re-printing a PDF through the print dialog only re-encodes a file that is already final.
If the source is a legacy binary .hwp file, the browser cannot read it at all; the HWPX to PDF guide explains that boundary.
A two-minute check before sending
Open the saved file in a viewer that is not the one that produced it — a phone works well. Confirm the page count matches your expectation, scan the first and last page, look for a table that was cut mid-row, and try selecting a paragraph of text: if nothing selects, you exported an image-only rendering and the file will not be searchable. Finally check the file name; browsers derive it from the page title, which is rarely what a recipient wants to receive.