Posted by qbow883 12/27/2025
On a separate note also not mentioned llm's are really good at generating ffmpeg commands. Just discuss with chatGPT your source file and goals for a video and you can typically oneshot a targeted command even if you aren't familiar with ffmpeg cli.
No, that's just nonsense for any guide targetting beginners, it's not fine, it's too error-prone and complicated and requires entering the whole unfriendly land of the cli!
> If you must use a GUI
Of course you must! It's much better to provide beginners with your presets in Handbrake that avoid the footguns you mention (or teach them how to avoid them on their own) rather than ask them to descend into the dark pit of ffmpeg "basics"
> Before you start complaining about how complicated ffmpeg is and how arcane its syntax is, do yourself a favor and read the start of its documentation. It turns out that reading the (f.) manual actually helps a lot!
It turns out that wrapping the bad UI in a simpler typed GUI interface wastes less of the collective time than asking everyone to read dozens of pages of documentation!
> Actual video coding formats are formats like H.264 (also known as AVC) or H.265 (also known as HEVC). Sometimes they're also called codecs, short for "encode, decode".
Codec is coder/decoder. It's not the format.
There's a footnote claiming people mix the 2 terms up (a video format is apparently equal to a video codec according to this "expert") but apparently acknowledging the difference is seemingly only what nitpickers do. Sheesh. If you want to educate, educate with precision, and don't spread your misinformation!
I would assert that the author was already being precise. A statement that X is "sometimes called" Y already conventionally carries the subtext that Y isn't actually the correct term; that Y is instead some kind of colloquial variant or corrupted layman's coinage for the more generally-agreed-upon term X.
Why mention the incorrect terminology Y at all, then?
Specifically in the case that pertains here, where far more laymen are already familiar with the Y term than the X term, giving Y as a synonym in the definition of X is a way to give people who are already familiar with this concept — but who only know it as Y — an immediate understanding of what is being discussed, by connecting their knowledge of "Y" over to the X term the author is defining. This is an extremely common practice in academic writing, especially in textbooks.
It's not fine to say "oh those 2 things are the same", especially in an introduction, because then you're leading people astray instead of educating them.
I think people of a certain age picked up that the things inside video containers were "codecs" when we all had to install "codec packs." The things inside those packs were literally codecs — encoder/decoder libraries.
But when AV software of the time (media players, metadata taggers, library software, format conversion software, etc) needed to let you choose which audio or video bitstream format to use for encoding [i.e. pick a codec plug-in library to pass the bitstream through]; or needed to tell you that you needed to install codec plug-in X to play file Y, they would often equivocate the codec with the AV bitstream format(s) they enabled the encoding/decoding of. Especially when the software also had a separate step or field that made reference to the media container format. It was not uncommon to see a converter with a chooser for "output format" and another chooser for "output codec" — where the "codec" was not a choice of what library plug-in to use, but a choice of which AV bitstream format to target. (Of course, relying on the assumption that you'd only ever have one such codec library installed for any given AV bitstream format.)
---
Heck, this is still true even today. Go open a media file in VLC and pop open the Media Information palette window. There's a "Codec Details" tab... and under each stream, the key labelled "Codec" has as its value the name of the AV bitstream format of the stream. I just opened an M4A file I had laying around, and it says "MPEG AAC audio (mp4a)" there.
My understanding of why VLC does it that way [despite probably decades of pedants pointing out this nitpick], is that there's just one codec library backing pretty much all of VLC's support for AV bitstream formats — "libavcodec". Players that rely upon plug-in codec libraries might instead have something like "MPEG AAC audio (libavcodec)" in this field. But since VLC only has the one codec, it's implicit.
Even though, to be pedantic, a media-info "Codec" field should just contain e.g. "libavcodec"; and then there should be a whole second field, subordinate to that, to describe which of the AV bitstream formats supported by the codec is being used.)
---
I also recall some old versions of iTunes using the word "Codec" to refer to AV bitstream formats in metadata and in ripping settings. Not sure if that memory is accurate; current versions seem to skirt the problematic nomenclature entirely by making up their own schema. Modern Music.app, on a song's "File" tab, gives the container format as the file's "kind", and then gives some weird hybrid of the AV bitstream format, the encode-time parameters, and the originating software, as an "encoded with" field.
But yes, as the other reply says, I am aware of this distinction, and I make a point not to use the word "codec" at any other point in the article. and explain in a lot of detail how much the encoder matters when it comes to encoding in some format. I mention the term to make people aware that it exists.
But, you're right, I will clarify this a bit more.