Calculate output size in advance, is that possible?

[Komet]

Hello everyone,

can you calculate the size of the output file? I always want to get a 9.9 GB file.

Convert from m2ts 4K UHD with HDR to MKV 4K UHD with HDR.

I have UniFab version 2.0.0.6

Example 1:
Input file size: 64.03GB, bitrate is set to 13736, audio format AAC 5.1. Output file: 9.83GB and UniFab shows me the size of the output file is 10.04GB.

Example2:
Input file: 86.58GB, bitrate is set to 11330, audio format AAC 5.1. Output file: 9.94GB and UniFab shows me the size of the output file is 10.62GB.

I actually only get the real output value after converting. I would like to have it beforehand.

Can I calculate the process in advance? I need a value that I can enter for the bit rate so that I can get 9.9GB in the output.

I have entered the same files into other programs and a different value comes out everywhere. Sometimes the output I see is 9.51GB before converting and a 10.81GB file comes out.

I wish you all a good start into the new year and thank you very much.

[Ryan_3]

Hi Komet,

Can I understand your question as follows: UniFab displays an estimated output file size that does not match the actual output size, and you want a more accurate estimate of the output size?

[Komet]

Hello Ryan_3

Yes, that is exactly my question. Can you calculate the output value exactly beforehand?

Before converting, UniFab always shows me an incorrect value in the file that comes out at the end.

Maybe you can use the parameters (size of the file, length of the film and size of the output file) to calculate the bit rate to be entered beforehand and which value I have to enter for the bit rate to get to approx. 9.9GB.

I myself did a lot of calculations beforehand and also looked a lot on the internet. But unfortunately everything was unsuccessful.

[raven-au]

I don't think that is possible because of the nature of the way video encoders work.
While the bitrate can be specified the video data itself is undetermined and the compressibility will be different for different video sources.
Therefore, the bitrate you specify is an average bitrate the algorithm will try and attain rather than a hard and fast bitrate since variability of the video source is unknown.
You should be able to get closer to your goal bitrate by using a 2 pass encode since it will use the first pass to better estimate the characterises of the video source in order to achieve the specified bitrate, that's partly what 2-pass encodes are for.

[Komet]

Hello Raven-au, it is not possible to sum up the final result. I created an Excel table and entered all the necessary data after each round. It always varies and is very difficult to achieve. Thanks for the information and the tip about the two passes, I'll follow up.

[raven-au]

What encoding settings are you using.
What I see is if I'm using 1 pass or 2 pass encoding is the calculated size is fairly close to what's produced, obviously if you use CRF encoding you won't get what it says is the calculated size because the bitrate is not used for CRF encodes, it's entirely determined by the CRF encoding value and the output cannot be determined.