Why does YouTube Export with Filters greatly increase file size

[codeowl]

Hi there,

I have a video taken on my iPhone 11 which is 94 mb. I loaded into Shotcut, did some editing cutting bits out, and detached the audio and removed it. I had a wav which was a mix of some audio I recorded separately at the same time as the vid which I then added. The wav file was 48 mb in size, and this also got cut down in the editing to match the vid.

I added a Fade In and Fade Out filter and exported this using YouTube option and the file size created is 50.4 mb. The video goes for 2:21 mins.

I then added the filters:

• Old Film - Scratches
• Old Film - Dust
• Old Film - Grain
• Saturation

I then exported this with the exact same YouTube setting, and the file size is 638 mb.
How can the file size be so radically bigger? It is like 12 times bigger.

What have I done wrong here?

Original Source Video File Spec:
Length: 3:40
Frame Width: 1280
Frame Height: 720
Data Rate: 3327 kbps
Total Bitrate: 3502 kbps
Frame Rate: 29.97 frames/second

Shotcut Project Video Spec:
Codec: H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10
Resolution: 720x1280
Frame Rate: 29.974125
Format: yuv420p
Aspect Ration: 9:16

Shotcut Export Setting:
Stock/YouTube
From: Timeline
Use hardware encoder = False
Format mp4
Resolution: 720x1280
Aspect Ration: 9:16
Frame/Sec: 29.974125
Scan mode: Progressive
Field order: none
Deinterlacer: YADIF - temporal + spatial (best)
Interpolation: Bilinear (good)
Use preview scaling = False
Parallel Processing = False

Thanks for your time.

Regards.

Scotty

[Elusien]

Don’t use the YouTube preset. Use the default preset instead. The reason why it produces such big files is to do with the GOP parameter explained here:

Size issue when exporting video

I have a problem exporting videos and I don’t know how to identify them or know what I did wrong. First export a 9gb video and a 1:30hs duration. Cut some parts remaining in 1:20hs and without editing anything else I export it with the “YouTube” profile because I assumed that it is lighter. The video was 1.10gb. The problem is that now I try to do exactly the same with another video (but with the same content, it is the recording of my screen of the same game, with the same characteristics, it…

[kagsundaram]

@codeowl

I then added the filters:

• Old Film - Scratches
• Old Film - Dust
• Old Film - Grain
• Saturation

Wow!
That is a LOT of video information added uniquely to each frame of your video.
Video size is a function of many things, one of which is the amount of difference between successive frames. These filters have added significant differences between each sequential frame. In order to maintain the Quality specified, the codec software must increase the percentage of I-Frames, correspondingly reducing the number of small, efficient P-Frames and B-Frames

@Elusien has explained this in another thread…

@Elusien

A GOP (Group Of Pictuires) always starts with an I-frame (Intra frame).
An I-frame contains all the information needed to produce the picture for that frame, it does not depend on the contents of any other frame. If the GOP consists of only I-frames it would emulate an old cinematographic spool of tape, where e.g. damage to one frame would not affect any other frame.

The GOP can also contain zero or more B-frames (Bi-directional frame).
B-frames consist of information that describes the changes in the picture between this frame and the previous one. So damage or corruption of an I-frame or B-frame would affect the validity of subsequent consecutive B-frames.

The GOP can also contain zero or more P-frames (Predictive frames).
P-frames consists of information that describes the changes in the picture beween this frame and the previous I-frame and also the following I-frame. In this way any discrepancies or corruption within previous B-frames are eliminated from this P-frame onwards.

…which it would do well to read in its entirety.

[brian]

@codeowl

I then exported this with the exact same YouTube setting, and the file size is 638 mb.
How can the file size be so radically bigger? It is like 12 times bigger.

I do not reproduce this with this specific filters. But I also do not have your exact same media files. I would ask you to carefully perform this test again. But this time, be very careful to only change adding/removing the filters. Be sure to choose exactly the same export preset with making any advanced changes.

If you can reliably reproduce the problem, then try to remove only one filter at a time until you find the one that causes the most change.

[codeowl]

Thanks to all that responded. It’s great to see such an active forum for this great product.
So I did the tests as suggested, and here are the results:
(Note Saturation was set to zero which is why it reduced the size a bit I guess)
(Also the only change I made for each test was to add filters, they were all done with the YouTube preset)
50.4 MB: FadeIn, FadeOut
45.3 MB: FadeIn, FadeOut, Saturation
46.6 MB: FadeIn, FadeOut, Saturation, OldFilm-Scratches
47.0 MB: FadeIn, FadeOut, Saturation, OldFilm-Scratches, OldFilm-Dust
638 MB: FadeIn, FadeOut, Saturation, OldFilm-Scratches, OldFilm-Dust, OldFilm-Grain

Then I did one more export with only the OldFilm-Grain filter:
674 MB: OldFilm-Grain

I only have two tracks in the timeline. One is an MP4, and the other is the wav. Only the MP4 has filters applied.

Regards,

Scotty

[shotcut]

This is as expected because grain adds a lot more information that cannot be removed without reducing the quality level to force it out.

[kagsundaram]

@shotcut

grain adds a lot more information

I lost my bet; I was betting on OldFilm-Dust.

But the principle is the same: lots of tiny bits of information, distributed uniquely across each frame.

[RilosVideos]

Its the same priciple as with JPG-compression or all other lossy compression algorithms: the more fine detail in the photo (or film) the bigger the file size. Adding fine grain noise is the worst you can probably do with respect to file size. Just be careful with those filters that add a lot of detail - at least be aware of what you are doing For a comparison: look at a photo with just a blue sky (or anything else with no detail). Then take a photo of grass, bushes with lots of fine detail. At the same jpg-compression rate you will have much bigger file sizes.