Shutter Encoder is an FFmpeg-based front end for transcoding, rewrapping, cutting, joining and changing media tracks. A job begins with an input list, one processing function and one or more destinations. Some functions create new compressed streams, while others copy compatible streams or change metadata. Selecting a function therefore matters more than the output extension alone.
Cutting without encoding follows keyframes
A compressed-video cut that avoids re-encoding snaps its boundaries to nearby keyframes. The output can begin or end away from the exact displayed frame. Frame-specific boundaries require a path that encodes affected material rather than assuming every frame can start an independent stream.
Conform changes frame-rate metadata. It does not repair variable-frame-rate timing. The documentation directs variable-frame-rate material to conversion instead. Rewrap changes the container without transcoding only when that container supports the existing streams.
Merge depends on matching sources
Merge requires the source files to match in codec, dimensions, frame rate and audio settings. Files with the same extension can still fail this test. A conversion step can normalize incompatible sources before another merge attempt, but that step creates new encoded media.
Replace Audio reads the video first in the input list and the replacement audio after it. Reversing that order changes which object Shutter Encoder treats as the base. Extract and replace operations also need the intended track selection when a source carries several audio or subtitle streams.
The queue can trade throughput for contention
The render queue can run jobs in parallel. The documentation recommends no more than two simultaneous video encodes. More parallel work can compete for CPU, storage bandwidth and memory rather than completing the batch sooner.
Hardware acceleration remains off by default because Shutter Encoder prioritizes CPU encoding quality. GPU decoding can also become slower when CPU filters remain in the chain and force transfers between devices. Enabling a hardware option therefore needs an output-quality and timing check with the actual function.
Watch folders wait for a stable file
A watch folder waits for a file to become available and checks its size at ten-second intervals before sending it to the configured function. This reduces the chance of processing a file while another program still writes it. A source that grows slowly or pauses between writes can still require a longer stability assumption outside Shutter Encoder.
With several output destinations, Shutter Encoder encodes the first output and copies that result to the other locations. It does not run a separate encode for each destination. A failed copy can therefore leave the main output complete while another destination remains missing.
Some operations need preparation outside the program
Screen-audio capture needs a separate virtual loopback audio driver. Installing Shutter Encoder alone does not create that operating-system audio route. Reverse holds the entire selected clip in memory, so the documentation recommends trimming the source first. These constraints come from the data path behind the function rather than the number of buttons in the interface.
