Both of these services significantly reduce the file-size with a re-encode, even if they promote an "impercievable quality loss". They seem to be more suited to high traffic on-demand streaming for websites, promotional material, etc.
FYI Mux isn’t an okayish platform for mid-level use-cases, it’s the gold standard for high-performance live streaming and VOD, created by the guys behind a bunch of other stuff pretty central to the streaming industry overall, and is used at the highest level by folks like CBS, Mandolin, etc. Of course you don’t have to use it, but it’s certainly no toy.
Just curious, does your audience expect to download it in full to watch later or on a USB stick as you describe, or is that a side effect due to the large file sizes?
EDIT: I am not a Mux employee, nor is any of this very important, I've just been working in this space for 10 years and very seldom run across something that needs these requirements and I'm curious :)
I encode H.264/AAC for compatibility, usually 2-pass at 8Mbit/s for FHD content or around 30Mbit/s for UHD, 256-320 kbps for AAC. This gives a size of 1-4 GB per video. My Dad worked in broadcast, gave me those figures 10 years ago, I generally stick by them for any work I'm proud of!
You are right, that bitrate absolutely isn't necessary :D , and there are better codecs too. I don't have more than 17 films at the moment, the whole backend costs me about 4 euros a month on Hetzner with an attached 32 GB of block storage, no bandwidth costs (300Mbit/s max), single-core running Node.js and Nginx.
I make films during travel and events, and share them with friends and family who were there and for who it has some intrinsic value. They're mostly personal, not just for consumption. Short compilations that I would like to keep for the future, like old photos. Hence why people watch (and hopefully rewatch!) them on a TV and don't mind the wait.
Buffer-less streaming is absolutely not a priority (although nowadays I think it's more expected, people have shorter attention spans, despite the hours of work that goes into a short travel film). It's a very niche case, but would have cost me at least $50 in bandwidth alone with the big three. It's not going to break the bank, but it's also unnecessary.
You don't usually notice the drop in quality on YouTube or Netflix, until you actually try working with high quality source footage (high-bitrate from a dedicated camera, not phone footage). Then it's very noticeable (purple/green flickering on even surfaces from chroma subsampling, I'm guessing), and makes you sad when you realise what everyone is missing!
If you're still curious, my website is still live. I beg you not to judge, I started making short films very young and never got very good at them either (I'm studying nanotech & engineering, no film career for me)!
For example, encoder might choose 1Mbps bitrate for a scene which is mostly static and 20Mbps for a part of the video with a lot of movement. Your 8Mbps constant bitrate will be an overkill for the first scene and too low for the second. Let encoder decide the optimal bitrate.
That doesn't mean they can't stream, there is an option to watch it in the browser, they just need a reliable connection (~30 Mbit/s).
