different PCs produce different encodes from the same source file?

[real_cheep]

Description of problem or question:
I used the latest version (1.3.3) to re-encode the same file on three different PCs, each running the latest version of Windows 10, just to see if I would end up with three identical output files, however, not only were they not identical, the difference in size between the smallest and largest was about 0.4%. I was very careful to set all the options the same for each encode, and did not use any GPU-based encoding. I had previously reasoned that if the encoder is an algorithm, albeit an extremely complex one, the same input should produce the same output, but obviously, this is not the case. What might account for the differences in the output files?

I also repeated the encode on one of the machines, and the output file differed in size from the previous attempt (on that same machine) by only three bytes.

[Deleted User 13735]

I was very careful to set all the options the same for each encode,

And that makes for a really imaginative story, but hardly gaspworthy. 0.4% is easily accounted for by differences in the headers from simply writing each datestamp and directory path!

If you will but post the required encode logs for all three encodes, it will be relatively easy for our colleagues to point out what is different in each; however, suffice it to say they are not "exactly" the same in the light of day, nor will they ever be.

For your future reference, you have employed a logical fallacy called Post hoc ergo propter hoc, meaning "If B occurred, then A must have caused it." It's a no-no in technical discussions.