Stephen Follows analyzes film industry data, covering box office, budgets, originality and how movies make money.
HN links it for data on Hollywood economics and originality, and for his post on TheNumbers.com and AI crawling. Commenters discuss its box office figures.
How much storage does Netflix actually need for its whole library?
It's got about 17,000 titles globally [1]. If they have copies in SD, 720p, HD and 4k that would be 68,000 versions (plus some extra audio tracks for stuff dubbed in multiple languages, but I suspect this is fairly minimal in terms of storage though)
Let's assume that the resolutions have the bitrates at 5, 10, 15 and 20 mbps.
The average length of a Netflix original movie is ~90mins [2]
So that would require about 575TB in storage if I have done my maths correctly.
You would need about 20x30TB Kioxia CD6 SSDs for all that. Very expensive but definitely technically possible.
I could totally see it being possible to fit those drives in a single node to push the 800gbps required, not increasing the over rack requirement at all. Not sure if the bandwidth from that many drives is enough, might have to cache some of the most watched stuff to ram)
Not gonna see any in home boxes with all the titles pre loaded any time soon though. As a hard drive array that's still 30x20TB drives.
[1] https://www.comparitech.com/blog/vpn-privacy/netflix-statistics-facts-figures/#:~:text=Netflix%20has%20over%2017%2C000%20titles,libraries%20as%20of%20October%202022
[2] https://stephenfollows.com/netflix-original-movies-shows/#:~:text=The%20average%20Netflix%20Original%20fiction,two%20shorter%20films%20mentioned%20above)
I was going to say this seems factually untrue, because the number of films released each year has been going up for a long time, but there are some interesting parallels on how films are distributed. Looking at https://stephenfollows.com/how-many-films-are-released-each-year/, the number of films released has gone way up over time, but the number of films in wide-release has gone done.
The internet has more content then ever, but the long tail of content there is consumed at a smaller rate relative to the big 3/5/100 web properties.
The problem isn't that consumer freedom has been killed- after all, all those films and content exist and likely more accessible than ever, its that the large centralized firms have managed to largely satisfy the majority of consumer demand without it. If I have the time to see a movie (a rare occurrence sadly) and Disney has a Marvel or Star Wars movie out, I'll just go see that, even if there might be a limited showing of a comedy-western-crime genre film out that I'd enjoy even more.
Note, I am concerned about media consolidation, but its more about distribution and IP ownership at this point than consumer choice.
Not GP, but looking at the relative difference graph[1] two hypotheses occur to me:
1) Most of the categories that are over-represented are ones that individuals watch out of obligation, rather than desire. A large portion of people watch documentaries (and biographies/historical films) because they feel they "should", because they feel like they should learn something.
Same for drama - these often include Oscar-bait that many people watch exclusively because it's nominated, not because they're actually interested in the movie itself.
Likewise, animation is (by volume) mostly synonymous with children's entertainment, so there are going to be a lot parents forced to watch what their kids are watching.
All of these line up with OP's hypothesis of obligation over choice.
2) War and westerns being overrepresented, well... film critics historically have a reputation of being Men of a Certain Social Demographic, and this is likely reflected even in modern film critic culture.
[1] https://stephenfollows.com/wp-content/uploads/2021/03/Difference-between-relative-Metascores-and-IMDb-user-scores-by-genre-2000-19-900x501.png
Have you checked that assertion? Because it didn’t ring true to me based on what I remember (which was that PG-13 tends to be the sweet spot), and it doesn’t match the data I can find, e.g.:
https://stephenfollows.com/which-mpaa-rating-earns-the-most-money/
PG-13 is far and away the leader in terms of both highest-grossing films and percentage of box office revenue; R-rated films are actually the second in terms of highest-grossing and, depending on the year, bounce around between second and third. G-rated films, by contrast, are at the bottom across the board.
There is a difference between “family entertainment” and “adults-only” which matters in terms of the discussion around Tumblr’s content. If we’re using MPAA ratings as an analogy, Tumblr’s “no porn” position isn’t “everything must be G-rated,” it’s “nothing can be NC-17.” One can argue about whether the latter is a good stance to take, but it’s important to be arguing about the actual restriction.
That list simply reflects the fact that older movies are forgotten and newer movies are more popular.
The rate of studio releases has been flat for decades and is even declining.[1] Television shows have certainly grown in number, diversity, and quality. But perhaps that's just because we spend more time in front of screens and have more disposable income, not because the pace of art and culture has sped up.
[1] See https://stephenfollows.com/how-many-films-are-released-each-year But before you argue that this is evidence of acceleration, if you read further down, and if you follow the money, it's more a reflection of a deluge of independent "releases" that nobody is actually watching. And it's not a long-tail phenomenon that's growing the viewership market; the vast majority of people only watch studio films.
Let me do a back of envelope calculation. This source [1] claims that the average production cost of a Hollywood movie has risen to $40 million. This is based on 5713 movies produced between 1999-2020, which is about 285 movies/year. Let's double that for fun and figure out what happens if you hit 500 million subscribers worldwide. 600 movies *40m/movie /500m subscribers = $48 /subscriber every year.
So presumably, only $4 dollars a month is more than enough to cover the cost of producing all Hollywood movies - ignore of course, all other sources of revenue such as cinema tickets and product placement etc.
Is marketing and distribution online really 9x the cost of producing the movie itself?
[1] https://stephenfollows.com/how-has-the-cost-of-making-a-movie-changed-over-the-past-twenty-years/