Computing desk
< August 22	<< Jul \| August \| Sep >>	August 24 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

August 23

Video file sizes

I recently downloaded an MKV video done with the h264 codec. It had video dimensions of 834 x 354, so even though it was a 2 GB file, the quality wasn't that good. So, I downloaded another version of nearly the exactly same file size and same codec, but with a screen resolution of 1280 x 544. Even without knowing the numbers, the video quality is clearly better. My question is: what was the poor quality video doing with its file size? Like, if it takes 2 GB to render a 1 1/2 movie into 720p, what was the other file doing with all those bytes that clearly won't being used on the screen? Yes, there were a few more subtitle streams in there, but remuxing them out shows they only account for a tiny sliver of the file size. What would account for the bloat? 70.27.163.9 (talk) 16:03, 23 August 2020 (UTC)[reply]

This really depends on your perspective.. An hour of raw (uncompressed) video, 1080p 30fps is 500 GB. If you think of that, then the difference between those two 2GB files isn't actually that much. Its much easier to 'waste' space than it is to save space. Video compression can be super efficient, but it also puts requirements on both the original material and the skill of the person applying the compression to make proper use of that efficiency. Because garbage in, means garbage out. —TheDJ (talk • contribs) 16:19, 23 August 2020 (UTC)[reply]

Yeah, encoders have a bunch of options and any idiot can do dumb stuff and make a ginormous, poor-quality file. 2 GB could easily be a raw, uncompressed 834-line video of something. Your thinking should rather be the other way around: amazement that high-quality video can be fit into only 2 GB. Modern video codecs do all kinds of black magic to cram more data into less space. --47.146.63.87 (talk) 06:29, 24 August 2020 (UTC)[reply]

Another thought that occurred to me is the first video could actually have been transcoded from another, less-space-saving codec, such as MPEG2. This is not uncommon. There is generally little point to this unless for some reason you need to play the video on a device that can only play one codec but not the other. The transcoded file will often be larger in size despite using a more-efficient codec, because the codec needs a high-quality video input to work its magic. But a lot of people don't understand this so they do it anyway. --47.146.63.87 (talk) 22:59, 24 August 2020 (UTC)[reply]