It is said, that UT Video is lossless.
So I have tried this:
Grabbed a Video from my usb-Grabber with “UT Video”.
After I have opened “UT Video”, moved it into the timeline section. I have not edited anything on it. Then I have exported it as “UT Video”.
My expectation was, that the size of the exported “UT Video” Video would be exactly the same like the original “UT Video”, because both are lossless. But they have different sizes. Why?
Thank you.
There are many variables.
Was the capture container format different than the export container? As in, if it was captured as an AVI file but exported as a MKV file, the files will be different sizes because data is organized differently even though it’s identical in terms of content.
There could also be a difference in chroma subsampling. Was it captured at 4:2:0 then exported as 4:2:2? The format itself is lossless, but a conversion would have happened if there is a subsampling difference.
Lastly, there’s the audio situation. Was audio captured and exported in the same format with the same settings like bit depth, sample rate, etc? Ut Video is only for video and has no effect on audio. So it’s the user’s responsibility to make sure audio settings are also set to lossless equivalents.
For capturing I took:
Shotcut->File->open other->device
Filter Export (Edit) ->lossless-> FFV1 UT Video (Edit) ->default settings
Button “capture file”
For exporting I took:
Shotcut->export tab->lossless-> FFV1 UT Video (Edit) ->default settings
Button “export file”
Unfortunately, this doesn’t help much. We would need Properties > drop menu > More Info on both the captured and exported clips to determine any differences between the files. That, or a MediaInfo probe result.
FFV1 is not Ut Video, so I’m unclear what’s happening.
In general, file sizes should never be expected to be identical. Containers have many optional metadata fields, and there’s no guarantee that an export process will create every optional field that a capture process did.
The two common ways to check for identical files are:
We also don’t know about subsampling yet. The capture device may have created 4:2:0 while the export may have created 4:2:2. The timeline is processed in 4:2:2, so simply invoking the timeline has potential to cause a chroma conversion if the source was 4:2:0.
I’m also unsure if “default settings” use the same audio settings for both capture and export.
Given the premise that file sizes should never be expected to be equal, is there an additional concern you have that the export is not truly lossless? Is there a visual quality problem that is causing concern?
Thank you.
There were some errors in my second post. So I have edited it.
In the meantime I have tested the same again.
And now the behaviour was as expected. The size of the second file has exactly the same size like the first. So maybe I did my first test wrong.
This topic was automatically closed after 90 days. New replies are no longer allowed.