A scan is the foundation of everything that comes after it — restoration, colourising, printing. Ten extra minutes of care here beats an hour of fixing later. Here's what actually matters: resolution, light, and the file format you archive in.
Before anything touches the scanner: dust off the print with a soft dry cloth, a lens blower or a clean make-up brush, and wipe the scanner glass. A speck of dust on the glass shows up as a bright white dot on every scan you make that day, and dust baked into a scan is one of the hardest things to remove afterwards. Never use water or cleaning fluid on the print itself — old photographic paper is easily damaged.
DPI (dots per inch) decides how much detail the scanner reads from the paper. More is not automatically better — past a certain point you're just magnifying paper grain and making enormous files. A practical rule of thumb:
| DPI | When to use it | What you get |
|---|---|---|
| 300 | Large prints (20×25 cm and up), or quick digitising of big boxes of photos | Fine for viewing on screens and same-size reprints; limited room to crop or enlarge |
| 600 | The default for standard prints (9×13, 10×15, 13×18 cm) | Enough detail to enlarge 2× and to give restoration software real information to work with |
| 1200 | Small prints, contact prints, passport photos, or when you'll crop out a single face later | Maximum honest detail from small originals; large files (often 50–200 MB as TIFF) |
A concrete example: a 10×15 cm print at 600 DPI gives roughly 2400×3600 pixels — plenty to re-print at the same size, crop generously, or run through an AI upscaler for a large-format print. The same photo at 300 DPI gives 1200×1800 pixels, which looks fine on a phone but runs out of detail the moment you crop in on a face.
One more tip: scan in colour mode even for black & white photos. A "black and white" print is rarely neutral — it's often warm-toned or yellowed — and a colour scan preserves that information, which makes later correction and colourising work better.
A modern phone camera can genuinely rival a mid-range flatbed — if you control the light. Most bad phone-scans fail for the same three reasons: glare, shadow, and perspective distortion. Here's the routine that avoids all three:
Dedicated scanning apps (Google PhotoScan, Microsoft Lens and similar) combine several angled shots to remove glare automatically, and they work well for glossy prints. For matte prints, a single careful photo in good light is usually just as good.
This is the step most people get wrong, and it's invisible until it's too late. JPG is a lossy format: every time a JPG is saved, the image is re-compressed and fine detail is thrown away permanently. Save a scan as JPG, open it, straighten it, save again, adjust the contrast, save again — and each save quietly degrades it. The damage shows up as blocky artefacts around edges, exactly where old photos hold their most fragile detail.
TIFF and PNG are lossless: the file can be opened, edited and re-saved a thousand times and remain pixel-identical. The practical workflow:
Yes, TIFF files are big — a 600 DPI scan of a 10×15 print runs 25–75 MB. Storage is cheap; your only surviving photo of a great-grandparent is not.
Even a perfect scan of a faded print is still a scan of a faded print. This is where restoration comes in: upload the scan to PicReviver and the AI rebuilds sharpness and lost detail — free, in your browser, and the file is deleted right after processing. It accepts TIFF and PNG directly (also JPG, HEIC and WEBP), so you can feed it your archival master at full quality. Blurry faces respond especially well to the dedicated face restore mode. And if you're wondering why the print faded in the first place, we wrote that up too: why old photos fade.