Working with frame rate conversion
The frame rate of a video asset represents how quickly video player devices play back the frames of a video, in frames per second (fps). For example, films have a frame rate of 24 fps, NTSC television broadcasts are 29.97/59.94 fps, and PAL television broadcasts are 50/25 fps. If you keep the MediaConvert default settings in your job, your output video will have the same frame rate as your input video.
Some videos have a frame rate that varies over the duration of the video. For example, some cameras automatically generate video that uses more frames for high-action sequences and fewer frames for sequences with less motion. MediaConvert supports variable frame rate (VFR) inputs, but creates only constant frame rate (CFR) outputs. For more information, see Variable frame rate inputs.
Topics
Settings for frame rate conversion
To create outputs that have a different frame rate than your input, use these MediaConvert settings:
- Frame rate (
framerateControl,framerateNumerator,framerateDenominator) -
For frame rate conversion, specify a frame rate in your output encoding settings that's different from your input video frame rate. MediaConvert will then create an output that has the frame rate you specify, rather than the frame rate of your input video.
Specifying your output frame rate directly in your JSON job specification can be more complex than doing so in the MediaConvert console. For details, see the procedure for using the API, CLI, and SDK in the topic Configuring frame rate conversion.
- Frame rate conversion algorithm
(
framerateConversionAlgorithm) -
Choose how you want MediaConvert to increase or decrease the frame rate. The best choice for this setting depends on the content of your video.
When you use Drop duplicate, MediaConvert copies or deletes frames but doesn't alter them. This preserves the picture quality of each individual frame, but might introduce stuttering in some conversions. For numerically simple conversions, such as 60 fps to 30 fps, Drop duplicate is often the best choice.
When you use Interpolate, MediaConvert blends frames together to avoid the need to repeat or remove frames. This results in smooth motion, but might introduce undesirable video artifacts. For numerically complex conversions, Interpolate is likely to provide better results than Drop duplicate.
When you use FrameFormer, MediaConvert uses the InSync FrameFormer library. The conversion uses motion-compensated interpolation based on the content of your input video. FrameFormer performs various frame rate conversion techniques on a scene-by-scene basis and can use different techniques on different regions of each frame. FrameFormer does the conversion based on automatic detection of your source video's underlying cadence, rather than relying on the frame rate reported in the file's metadata.
Feature limitations:
-
You can use FrameFormer with inputs that have resolutions up to 4K only. MediaConvert doesn't support FrameFormer conversion with 8K inputs.
-
You can use FrameFormer only with jobs that you run through an on-demand queue. You can't use reserved queues with FrameFormer.
Using FrameFormer increases the transcoding time and incurs a significant add-on cost. For more information, see the MediaConvert pricing page
. -
Depending on the conversion, you might also use these settings:
- Slow PAL (
slowPal) -
When you convert the frame rate from 23.976 or 24 frames per second (fps) to 25 fps, you can optionally enable Slow PAL (slow phase alternating line ). When you enable slow PAL, instead of duplicating frames to increase the frame rate, MediaConvert relabels the video frames as 25 fps and resamples your audio to keep it synchronized with the video. Slow PAL frame rate conversion slightly reduces the duration of the video. Generally, you use slow PAL to convert a cinema format for file-based playback or internet streaming.
- Telecine (
telecine) -
When you convert the frame rate from 23.976 frames per second (fps) to 29.97 fps, and your output scan type is interlaced, you can optionally set Telecine to Hard or Soft to create a smoother picture. Generally, you use telecine when you're preparing a video asset for broadcasting to set-top boxes.
For more information, see Telecine.
- Scan type (
inputScanType) -
Use this setting only with progressive segmented frame (PsF) inputs. MediaConvert automatically detects progressive and interlaced inputs. But it doesn't detect PsF. When your input is PsF, set Scan type to PsF for better preservation of quality when you do deinterlacing and frame rate conversion.
