Quality loss after export

Thank you.

I made sure that all three images above are from the exact same frame.

Even when I compare both full frames (without enlarging) it is such a difference.

It is clear during the playback that the entire film has a much lower quality after the export.

(I enlarged the areas to make it more visible how much detail is lost.)

So is there anything I could do?

If it would help I could supply more frames for comparison.

Thank you Luis.

The source material are Capto screen recordings.

(Before using Capto I tried ScreenFlow, but the performance was inferior.)

By "optimizing" – do you mean importing while FCP creates these large files for easier editing?

I did this years ago, but singe Macs became more powerful it was no longer required.

So in this case the original 1.22GB file (and other similar ones) are just referenced.

Or has optimizing other advantages which I don't know?

Here's the information you had asked for (from QT & Invisor):

1. Before trying optimized files, I searches for official information and could only find this:

• Create optimized media: This option transcodes video to the Apple ProRes 422 format, which provides better performance during editing, faster render times, and better color quality for compositing. If the original camera format can be edited with good performance, this option is dimmed.

In this description there is no hint that it would improve the quality of export. Why do you think that it could?

2. Regarding variable frame rate:

I once used HandBrake (I still have version Version 1.0.0 (2016122400)/Mac App Store installed), but found it difficult to use (which options to use or better not to use was always the question). So unfortunately I would need more help how this would work.

On the other hand, I already use the following (via Terminal) to remove doubled frames (which sometimes occur) from screen recordings:

ffmpeg -i in.mov -vf mpdecimate,setpts=N/FRAME_RATE/TB out.mov

Probably there is just an additional parameter required to fix the frame rate?

3. Yes, you are right: foliage (like sea/waves) is difficult for screen recording. Additionally it is a fast tracking shot! (camera follows the tracks) Therefore the data rate is so high, probably also because the screen recording should not use a lot of the cpu power just for higher compression – which could have an impact on the 3D graphics rendering app which works at the same time.

Thanks for your answer.

> Transcoding never “improves” an image.

Yes, I didn't mean that, I just wondered if it could improve the end result. I guess I will have to try it.

I'm still trying to find out what's the combined ffmpeg code for removing doubled frames and to get a fixed frame rate. I learned that for changing the frame rate the video would have to be reprocessed.

> you probably need to export with a higher data rate

What would you say is the maximum rate which is still streamable?

I have one question regarding Optimizing:

My screen recordings are already very large files, and the resolution is 2880 × 1620.

My FCP project however is 1080p.

After optimizing theses files, will they be reduced to 1080p or stay 1620?

Thanks.

... and here is the source frame (Capto, 1620p) as jpg (couldn't post the PNG, was > 5MB).

It looks much better where the attention is drawn to – where the tracks are going: