How to use
- Enter the original width in pixels, e.g. 1920.
- Enter the original height in pixels, e.g. 1080.
- Enter the scale factor: 2 doubles each side, 0.5 halves it, 1.5 enlarges by 50%.
- Press Convert. The new size appears as width × height.
- To hit a specific target width, compute the factor first: target ÷ original (1280 ÷ 1920 = 0.6667).
- If the output shows an exact value in brackets, the size was rounded to even pixels.
Formula / how it works
The tool applies one factor to both dimensions and rounds each to the nearest whole pixel:
new width = round(width × scale)
new height = round(height × scale)
Because the same factor is used on both sides, the aspect ratio (width ÷ height) is preserved up to rounding. Pixels cannot be fractional, so JavaScript's Math.round is applied separately to each side — which is why some results drift by a pixel from the "ideal" ratio.
Total pixel count grows with the square of the factor. A scale of 2 doubles width and height but quadruples the number of pixels: 1920 × 1080 is about 2.07 megapixels, 3840 × 2160 is about 8.29 megapixels. File sizes, GPU memory and render times rise roughly in proportion to pixel count, not to the linear factor.
Common aspect ratios and sizes:
- 16:9 — 1280 × 720 (HD), 1920 × 1080 (Full HD), 2560 × 1440 (QHD), 3840 × 2160 (4K UHD)
- 16:10 — 1280 × 800, 1920 × 1200, 2560 × 1600 (many laptops)
- 4:3 — 1024 × 768, 1600 × 1200
- 1366 × 768 — the old budget-laptop panel; it is close to, but not exactly, 16:9 (ratio 1.7786 versus 1.7778)
Even-pixel rounding. Most video codecs (H.264, H.265) require even dimensions because 4:2:0 chroma subsampling stores colour at half resolution, and FFmpeg rejects odd sizes for these codecs. The tool therefore rounds each side to the nearest even number: 2 × round(exact ÷ 2). When rounding was needed it also prints the exact product so you can see what changed. 1366 × 1.5 = 2049 exactly, which becomes 2050. Invalid input (zero, negative or blank) returns an error instead of a size.
Upscaling does not add detail. Doubling a 1080p frame to 2160p yields a 4K-sized file, but the extra pixels are interpolated (bilinear, bicubic, Lanczos or AI upscalers). Text and edges may look softer than native 4K footage.
Device pixel ratio. On high-density screens, CSS pixels differ from physical pixels. An image shown at 400 × 300 CSS px on a 2× display needs an 800 × 600 source to look sharp — exactly the scale-2 calculation this tool performs.
Social media and web slots. Platforms resize uploads to fixed widths, so preparing the right size avoids a second, lossy resample. Instagram feed posts display at 1080 px wide; a 4:5 portrait is 1080 × 1350. YouTube thumbnails are 1280 × 720. Open Graph link previews look best at 1200 × 630. If your source photo is 4032 × 3024 from a phone camera, a factor of 0.2679 brings the width to 1080 and the height to 810 — a 4:3 frame you would then crop.
Input checks. The tool does not stop you entering zero or negative values; it simply multiplies them. Use positive integers for real images.
Worked examples
Full HD to 4K UHD
Inputs: Width 1920, Height 1080, Scale 2
Enter 1920, 1080 and a scale of 2. The tool computes round(1920 × 2) and round(1080 × 2).
Result: 3840 × 2160
Four times the pixels: 8.29 MP instead of 2.07 MP.
Scaling a 1080p frame down to 720p
Inputs: Width 1920, Height 1080, Scale 0.6667
Target width 1280 ÷ 1920 = 0.6667. Enter that scale; the tool gives 1920 × 0.6667 = 1280.06 and 1080 × 0.6667 = 720.04, then rounds both to even pixels.
Result: 1280 × 720 (exact 1280.06 × 720.04, rounded to even pixels)
Using 0.67 instead would give 1286 × 724, so keep four decimals.
Old laptop panel at 150%
Inputs: Width 1366, Height 768, Scale 1.5
Enter 1366 × 768 with scale 1.5. 1366 × 1.5 = 2049 exactly and 768 × 1.5 = 1152; 2049 is odd, so it becomes 2050.
Result: 2050 × 1152 (exact 2049 × 1152, rounded to even pixels)
The aspect ratio shifts by under 0.05%, invisible in practice.
Use cases
- Planning 2× and 3× image assets for retina phones and laptops.
- Working out export sizes when upscaling or downscaling video.
- Shrinking screenshots to a consistent width for documentation or blog posts.
- Estimating how many pixels (and roughly how much memory) a larger render will need.
- Checking whether a resized image still matches a 16:9 or 4:3 slot.
Tips & limitations
- Factor = target width ÷ original width; use at least four decimals.
- Round video dimensions to even numbers.
- Pixel count scales with factor², so 2× is 4× the data.
- Upscaled images look softer than native resolution.
- For CSS, multiply display size by devicePixelRatio to get the source size.
FAQ
Does the converter keep the aspect ratio?
Yes. It multiplies width and height by the same factor, so the ratio stays the same apart from rounding to whole pixels.
How do I scale to an exact width?
Divide the target width by the original width and use that as the factor. For 1920 → 1280, the factor is 0.6667.
Why is the result slightly different from width × scale?
Each side is rounded to the nearest even pixel so the size works with H.264/H.265 video. The exact product is shown in brackets whenever rounding happened.
Is 1366 × 768 a true 16:9 resolution?
Not exactly. 1366 ÷ 768 ≈ 1.7786, while 16 ÷ 9 ≈ 1.7778. It was chosen to fit memory sizes and is effectively 16:9 in practice.
Does doubling the resolution double the file size?
Roughly quadruples the raw pixel data. Compressed file sizes depend on content and codec but usually grow by 2.5–4×.
Can I convert DPI or print sizes here?
No. For print, inches = pixels ÷ DPI; a 3000-pixel-wide image at 300 DPI prints 10 inches wide.
Results are calculated from the values you enter. Double-check measurements and allow a margin before buying materials or making decisions.