How Do You Choose Video Aspect Ratios and Formats?
Whether a video is shot horizontal or vertical decides what fits in the frame before any platform comes into it. Ratio looks like a setting applied to a file afterwards, but it is usually a decision made on shoot day.
Ratio is shape, resolution is detail
Aspect ratio is a picture's width divided by its height. 16:9 means 9 units of height for every 16 of width. Resolution describes how many pixels draw that same shape. The 1080p label points at the height. A vertical video at 1080×1920 has the same total pixel count as a horizontal one at 1920×1080. What separates them is shape, not quality.
That also sets the order of decisions. The ratio is chosen first, then the resolution to deliver at in that ratio. YouTube's help page lists 1080p as 1920×1080 and 4K as 3840×2160 for the default 16:9 ratio. For a video uploaded in another ratio, the player adapts itself to the size of the video and the viewer's device.
Three ratios leave three different areas in the frame
Shooting one subject in three ratios opens three different windows onto the same scene. The horizontal window carries things standing side by side. The vertical window carries things stacked. The square sits between them and leaves no spare room at the edges.
| Ratio | Example size | What sits comfortably in frame | Where to be careful |
|---|---|---|---|
| 16:9 | 1920×1080 | People side by side, wide locations, a product with its surroundings. YouTube's standard ratio on a computer | On a phone held upright it shows as a small strip |
| 9:16 | 1080×1920 | One person, a face shot, a product explained top to bottom. YouTube describes Shorts uploads as vertical video | Two subjects side by side do not fit. Interface elements at top and bottom can cover text |
| 1:1 | 1080×1080 | A single subject at the centre, a product detail, a short text card | Neither fully horizontal nor fully vertical, so it leaves slack in neither direction |
What cutting three ratios from one shot costs
If you take a vertical 9:16 slice out of a horizontal 1920×1080 frame and keep the full height, only 608 pixels of width remain. That is about a third of the frame's width. A square slice keeps 1080 pixels, a little over half. Whatever lies outside the slice is thrown away. If it sat at the edge of the scene, nothing is lost. If two people stand in the middle of the scene, one is cut.
Done before the shoot, this arithmetic opens two routes. One is to keep the subject in the centre and build the horizontal frame around the vertical slice. The other is to shoot vertical and horizontal separately. The second asks for more shooting time. The first leaves empty space in the horizontal version. Both are choices, and both belong in the shot list. What is to be said, and which frame stands in which scene, is covered in the piece on why a shot list is not enough.
Fictional example
This example is not a client record. In a promo shot for a kitchen tool, the tool sits on the left of the table and the person using it on the right. The horizontal frame carries both. When the slice moves to vertical, only one of them fits. If the tool and the person had been brought toward the centre during the shoot, a separate shoot for the third format would not have been needed.
Adding black bars is not a fix
Putting black strips above and below a horizontal video to fit a vertical screen builds the problem into the picture. YouTube's help page advises against adding padding or black bars directly to the video, because the bars get in the way of the player adapting to the video's size and the device. When a vertical video is watched on a computer, the player may add its own padding. It is white by default and dark grey in dark theme.
The takeaway is that the file should be delivered in the ratio it will be shown in. If the target is vertical, the edit is vertical. Carrying one horizontal file everywhere with padding is not a video prepared for the place it appears.
Compression and the delivery file
Compression carries the same picture with less data. As the bitrate drops, the file shrinks, and fine detail and fast motion degrade first. For uploads, YouTube recommends the H.264 video codec, an MP4 container, and AAC-LC or Opus audio. The frame rate should stay the same as the recorded frame rate. Interlaced content should be deinterlaced before upload.
The bitrates below are the reference values YouTube gives for SDR uploads. They are recommendations, not hard limits. Another platform's values may differ.
| Resolution | 24, 25, 30 fps | 48, 50, 60 fps |
|---|---|---|
| 1080p | 8 Mbps | 12 Mbps |
| 1440p (2K) | 16 Mbps | 24 Mbps |
| 2160p (4K) | 35–45 Mbps | 53–68 Mbps |
Before delivery
If the questions below have no answer, the ratio decision is not finished. They run from the shoot to delivery.
- Which screen the video will be watched on, and whether that screen is held horizontal or vertical.
- If several ratios will come from one shot, whether the slice arithmetic was done before the shoot.
- Whether the key subject was confirmed in the shoot frame to stay inside the central slice in every ratio.
- Whether text and logo sit in each ratio's safe area. Platform interface can take space at top and bottom.
- Whether the file is in the ratio it will be shown in, with no black bars or padding baked into the video.
- Whether the frame rate matches the recorded frame rate, and interlaced content was deinterlaced.
- Whether the target platform's current size and file page was checked.
Sources
- Video resolution & aspect ratios
Accessed October 20, 2026
- YouTube recommended upload encoding settings
Accessed October 20, 2026
- Get started creating YouTube Shorts
Accessed October 20, 2026