Desqueezing restores the proportions compressed by an anamorphic lens: for a horizontal squeeze, final display aspect ratio equals recorded width ÷ height × squeeze factor. Calculate the corrected resolution, delivery crop and timeline settings from exact pixels or just a recording aspect ratio, including portrait footage and rotated lenses.
Last reviewed: · Methodology and reproducible checks below.
Footage and lens inputs
🔒Calculations run locally in your browser. This tool does not upload footage or entered values.
Desqueeze results
Native desqueezed aspect ratio—
Nominal desqueezed canvas—
Rounded canvas suggestion—
Upright recorded ratio—
Geometry correction—
Retained image dimensions—
Original image retained—
Crop analysisEnter valid values to calculate.
The diagrams compare framing before delivery resizing. The final frame label shows the delivery dimensions when a timeline is set.
Apply in your editor
Open Timeline settings to enter a delivery size.
Scale percentages use a 1:1 pixel baseline with automatic fit disabled. Apply desqueeze once. Editors with automatic scaling or interpreted pixel aspect ratios may show different transform values. Coordinates use pixel boundaries from the top-left; fractional boundaries are intentional.
How to calculate an anamorphic desqueeze
Choose exact stored pixel dimensions or a recording shape. Shape buttons provide starting points; enter your actual raster for precise pixel results.
Select the marked squeeze factor. 1.33× and exact 4:3 are deliberately separate.
Choose the delivery ratio. Open Camera orientation for portrait files or a lens rotated 90°.
Open Timeline settings to choose crop-to-fill or fit-with-bars and an optional final-frame width or exact delivery resolution.
Read the comparison and editor settings, then copy or download the report. Advanced options contains custom ratios and rounding.
A recording mode’s name may only approximate its pixel ratio. For example, ARRI lists ALEXA 35 “3.3K 6:5” as 3328 × 2790: use those pixels when calculating that specific mode. Some camera formats are already desqueezed; do not desqueeze those twice. See the ARRI ALEXA 35 workflow guide, format table.
Calculation methodology and reproducible checks
Let W and H be stored dimensions and S the squeeze. Correct the selected file axis, then rotate if needed. “Expand” multiplies that axis by S; “keep” divides the other axis by S. Both conventions give the same proportions. A 90° camera rotation swaps the corrected width and height.
Horizontal: Wd = W × S; Hd = H; ratio = (W ÷ H) × S
Vertical: Wd = W; Hd = H × S; ratio = (W ÷ H) ÷ S
Keep horizontal axis: Wd = W; Hd = H ÷ S
Keep vertical axis: Wd = W ÷ S; Hd = H
After 90° rotation: swap Wd and Hd
For delivery ratio T, the largest centred crop is Cw = min(Wd, Hd × T), Ch = min(Hd, Wd ÷ T). Each side loses (Wd − Cw) ÷ 2; each top/bottom edge loses (Hd − Ch) ÷ 2. Retained area is Cw × Ch ÷ (Wd × Hd). For fit, retain the entire original image and add bars instead.
Timeline Tw × Th (square pixels):
Crop-to-fill scale k = max(Tw ÷ Wd, Th ÷ Hd)
Fit-with-bars scale k = min(Tw ÷ Wd, Th ÷ Hd)
Clip top-left = ((Tw − k × Wd) ÷ 2, (Th − k × Hd) ÷ 2)
Fill crop size = (Tw ÷ k, Th ÷ k)
Final pixel count = Tw × Th (includes bars for fit)
Timeline settings use the actual integer raster, so derived-height rounding can change the effective ratio slightly. Nominal values are calculated without intermediate rounding; displayed decimals are approximations. Source crop coordinates undo the axis scale and, where selected, the 90° clockwise rotation. They describe a centred boundary rectangle, not inclusive pixel indices.
Check 1: 3072 × 2560 with 2× horizontal expansion → 6144 × 2560 → 2.4:1. At 3840 × 1600, k = 0.625: 62.5% corrected-canvas scale, raw X = 125%, raw Y = 62.5%; zero crop; 6,144,000 delivery pixels.
Check 3: Fit the first canvas into 1920 × 1080: k = 1920 ÷ 6144 = 0.3125; image = 1920 × 800; bars = (1080 − 800) ÷ 2 = 140 px top and bottom; retention = 100%.
Worked examples — load a scenario
These are illustrative rasters, not promises of a specific camera mode. Each scenario loads exact pixels, horizontal expansion and crop-to-fill; the portrait examples also set rotation and lens axis. Arithmetic uses nominal lens factors.
The 6:5 / 2× relationship follows directly from 1.2 × 2 = 2.4. For a 2.39 delivery the ideal stored ratio is 2.39 ÷ 2 = 1.195; ARRI explains why this is approximately 6:5. Actual camera rasters can differ from their mode names.
Apply the result in an editor or monitor
Use a square-pixel timeline at the displayed delivery dimensions. First confirm whether the file is still squeezed and whether metadata already corrects it. Either interpret matching pixel-aspect metadata or apply the axis transform manually, then scale and centre for fill or fit. Do not combine two desqueeze corrections. The calculated percentages are mathematical transforms from original pixels; disable automatic scaling when applying them directly.
DaVinci Resolve
Check the clip’s interpretation in Media Pool → Clip Attributes → Video → Pixel Aspect Ratio. Use the matching desqueeze when available; otherwise use independent X/Y sizing. Set the timeline to the calculated raster and use centred fill or fit sizing. For direct percentages, select a 1:1 / centre-crop input scaling baseline and apply the shown transform (Resolve zoom factors are percentages ÷ 100). Check a frame leader before export. Blackmagic’s reference manual documents clip interpretation and input scaling; ARRI’s finishing workflow illustrates metadata checks and Resolve output sizing. Menu wording varies by version.
Adobe Premiere Pro
Use Clip → Modify → Interpret Footage to inspect or override pixel aspect ratio when a preset matches the actual lens factor. If it does not, keep square-pixel interpretation and use independent horizontal/vertical Motion scaling. Start without Scale to Frame Size or other automatic fitting when using the raw X/Y percentages. Centre the image in a square-pixel sequence at the calculated timeline size. Adobe: correcting pixel-aspect interpretation.
Final Cut Pro
Set the project’s custom resolution, then use Video inspector → Spatial Conform → None for a 1:1 starting point. Expand Transform → Scale to adjust X and Y separately using the raw percentages. For already corrected square-pixel media, Spatial Conform Fit preserves the image with bars and Fill crops the edges. Apple’s Anamorphic Override is documented for SD widescreen, so it is not a general 1.8× or 2× lens selector. See Apple’s frame conforming, Transform controls and metadata override instructions.
On-camera and external monitors
Select the lens’s desqueeze factor and add a delivery frame guide to judge composition. A viewing correction does not establish whether the recorded file is corrected: check the camera’s recording mode and any HDMI/SDI output processing. Avoid a second monitor desqueeze on an already corrected feed. Rotated or vertical-axis lenses require compatible axis/rotation controls. Atomos documents desqueeze monitoring and crop frame guides.
Assumptions and production limits
This is a uniform geometric model with square-pixel source sampling, a selected squeeze axis, optional 90° rotation and centred framing. It does not model lens distortion, focus-dependent squeeze, anamorphic mumps, stabilisation margins, non-square source sampling or off-centre reframing. A nominal lens factor is a planning value; verify critical material with a known circle or frame leader.
Expanding a raster does not create captured detail. Confirm codec dimension limits and alignment, especially for odd-sized custom outputs. Ratio-only mode cannot determine source pixels, source crop coordinates or clip scaling percentages. A delivery width supplies output pixels only. Camera metadata and editor automatic scaling may change the meaning of an editor’s 100% scale.
Anamorphic desqueeze FAQ
How do I calculate a desqueezed anamorphic aspect ratio?
For a horizontal squeeze, divide stored width by height and multiply by the squeeze factor. A 6:5 recording with 2× becomes 2.4:1. For a vertical squeeze, divide the recorded ratio by the factor. Rotating the corrected frame 90° takes the reciprocal of its ratio.
Should I multiply width or divide height to desqueeze?
For horizontal squeeze, either multiply width by the factor or divide height by it. Both correct the same proportions, with different nominal raster sizes. For vertical squeeze, multiply height or divide width instead.
Does a 1.33× lens on 16:9 make exactly 2.39:1?
No. (16 ÷ 9) × 1.33 = 2.364444…:1. Exact 4:3 squeeze gives 2.370370…:1. Both need a small top-and-bottom crop to fill 2.39:1.
Why can the exact desqueezed dimensions be fractional?
A decimal squeeze factor can create fractional nominal dimensions. Transforms can retain these proportions internally; an exported raster needs integer dimensions. This calculator shows nominal values and separately rounds derived output sizes.
Is the marked lens squeeze always exact?
Not necessarily. Effective squeeze can vary with lens design and focus distance. Treat the marked factor as a planning value and verify critical footage with a known circle or square.
Should I crop to fill or use letterboxing?
Crop-to-fill covers the delivery frame and removes image at opposing edges. Fit preserves the whole image and adds top/bottom letterboxing or left/right pillarboxing. Choose fit when preserving composition matters more than filling the screen; choose crop when the delivery must fill its frame.
What is the difference between 2.39:1 and 2.40:1?
They are close but distinct mathematical ratios. At 3840 pixels wide, the nominal heights are about 1606.695 and 1600 pixels respectively. Cinema scope uses specified rasters such as 2048 × 858 or 4096 × 1716, whose ratio is about 2.386946:1. Use the required delivery raster rather than inferring it from a rounded name.
What is the best sensor shape for a 2× anamorphic lens?
For horizontal 2× squeeze and 2.40:1 delivery, 6:5 is an exact mathematical match. For 2.39:1, the ideal recorded ratio is 1.195:1, approximately 6:5. A 4:3 recording becomes 2.666667:1 and offers side-crop room. The practical choice also depends on the camera’s actual raster, lens coverage and reframing needs.
Should I deliver squeezed footage or square-pixel files?
Follow the receiving pipeline’s specification. Squeezed originals can be appropriate for camera-original or VFX handoff when squeeze metadata and framing instructions accompany them. A corrected square-pixel export is usually easier for general playback. Confirm the required raster and pixel aspect ratio; do not apply desqueeze twice.
Does monitor-only desqueeze change my recorded file?
A monitor viewing correction alone does not prove the file has been resampled. Cameras and recorders can have separate recording and output settings. Inspect the stored raster and metadata, and test a clip in your editor before deciding whether another correction is needed.
Can I use the calculator for anamorphic photos?
Yes. The same geometry applies to a still image. Start from its stored pixel dimensions, apply the appropriate horizontal or vertical resize with proportions unlocked, then crop or fit for the intended print or export. Expanding dimensions does not add captured detail.
How does vertical anamorphic work?
Camera rotation and lens rotation are separate choices. Rotating the camera 90° swaps the frame dimensions and turns a normal horizontal file-axis squeeze into a vertical squeeze in the upright image. Rotating the lens 90° changes the squeeze axis relative to the sensor. Set both controls to match your footage, then choose 9:16, 4:5, 1:1 or 3:4 delivery.
Can I plan with only a recording aspect ratio?
Yes. Ratio mode calculates the desqueezed shape and retained area without inventing source pixels. Add a final-frame width or exact delivery resolution for an output timeline. Enter exact source dimensions when you need clip scaling percentages or source-pixel crop coordinates.
Does this tool upload my footage or inputs?
No. The calculator works locally from the numbers entered. It does not request media files or send production values to a backend.
