First, let me state that this is not some sort of apologia for AAA and its business practices. I don't work for an AAA publisher. I don't derive significant annual revenue from AAA. I am currently an independent designer and consultant, but I have worked in PC gaming, MMO, console, mobile, and social web games over the last couple of decades. I have been writing about games and game industry issues for just about all of that time, and in that time have both attacked and defended many different industry practices. Among other things, I've attacked lootboxes, addictive mechanics, and the ethics of game service operators many many times. Your fight is not with me, and impugning my integrity is uncalled for. Skepticism, yes, of course. But there's no need to accuse me of dishonesty.
To address specific points:
Cost to users per unit sold is not graphed, but it's in the dataset. I had to go find it per title by hand. This is pretty easy; the vast majority of the points on the graph are $50 or $60. Free to play games are excluded from the player price paid for MB because there isn't anywhere enough data out there on LTV for F2P titles. MMOs from when they were in the boom period were baselined at $150, which matches my personal industry experience from that time, and is actually probably a little under.
Cost to publishers per unit sold is simply not available. Units sold for most games aren't available, unfortunately.
Revenue per unit sold is even more confidential than dev costs, and I don't have any figures at all for it. It's very rare for it to come out except in segments that use ARPU, and even then ARPU tends to be baselined on per day not lifetime revenue. Investor calls tend to only announce total revenues, not broken out by title.
"Value of game to a consumer" is utterly subjective, of course. I make no pretense at assessing that. You might highly value Mortal Kombat games; I don't tend to like them. I think Edith Finch was one of the best experiences of the year, you might prefer Zelda or Odyssey or Nidhogg 2. At best we could try for something like hours played, but that's an even less available figure.
So yeah... bytes are a very rough proxy, and far from an ideal one. They are just a (perhaps poor) stand-in for work product. They go some ways towards representing the amount of work that goes into a game these days, which unfortunately many game players seem to be unaware of. They are not a perfect proxy by any means. But part of the point in using them is that we can then look at a highly systemic game with low art costs that still did very well (like FTL, around 180MB, as you mention) and see that it is an outlier on the curve.
It is, unfortunately, incorrect that presence of absence of bytes has no value to a consumer. It's certainly not any sort of direct relationship, and it's not universal. But I think it's pretty fair to say that for a content-driven game, bytes are a pretty noticeable value. Graphics; levels and maps; characters and character choices; number of enemies... these are all things that players can and do use to measure the relative merit of content-driven games and we have all seen it. Not a universal, but common enough that it would be foolhardy to completely dismiss it. Byte expenditure is in many cases basically a marketing tool, as it contributes to art direction in particular. Cool visuals get you noticed. Do cool visuals always correlate to high byte count? No, and that would actually be a trend we would (and do) see as developers seek out cheaper yet well art-directed visuals.
As far as EA or other AAA shops. You can actually go look, they have in fact given percentage figures for incremental digital revenue versus retail, in some investor calls. The key thing is mostly that digital revenue is rising year on year. There is absolutely a debate to be had about "what margin should a publisher make" and "how much upselling is appropriate given that I already paid a retail cost." That's not what this article was about, though. I make no pretense to speak about profitability right now in this article. The point of it is the trendlines.
If you don't have costs, do you really have anything resembling an accurate picture of the financial side of the game industry? Cost is pretty much second only to revenue in terms of its necessity to discuss the health of a business.
> bytes are a very rough proxy, and far from an ideal one
I'd go so far as to say it's a useless proxy, frankly. Games in the 80's, 90's, and early 00's were externally limited in the number of bytes that could even be shipped to the customer, so of course it's going to go up with time, even if the effort that goes into producing those bytes were to trend to 0.
> The point of it is the trendlines
So, perhaps we should instead be focusing on how to keep the cost trend-line in line with the available funds trendline, without sacrificing a whole categories of games or finding ways to artificially (and IMO temporarily) increasing the available funds trendline?
After all, to borrow some examples from another thread between us, there are only two successful soccer games at this point in time. Three-ish competitive first person shooters. Two Battle Royale shooters. No real successful chess games (except for Battle Chess. Battle chess rocks). Trying to break into those kinds of long-term entrenched markets is going to be a nightmare, even when you have a good game.
In comparison, how many major contenders for a player's wallet exist for a new choose-your-own-adventure style game? For a open world game with an engaging characters and re-imagined open world tropes? For a new hybrid real time strategy slash turn based strategy game with a kick-ass soundtrack? None. Even the turn-based fantasy RPG market has started to show a vacuum a mere 4 months after the release of "Divinity 2".
Your examples miss my point, which was that some types of games are inherently more expensive to make than others.
Just like some types of games inherently retain better than others, generate more hours played than others, are more competitive than others, appeal to broader audiences than others and so on. They are can all be good games and still vary tremendously in this way.
All I was saying was that therefore the market dynamics will push towards game design approaches that help with the cost problem, and that spells trouble for game design approaches that don't.
First, let me state that this is not some sort of apologia for AAA and its business practices. I don't work for an AAA publisher. I don't derive significant annual revenue from AAA. I am currently an independent designer and consultant, but I have worked in PC gaming, MMO, console, mobile, and social web games over the last couple of decades. I have been writing about games and game industry issues for just about all of that time, and in that time have both attacked and defended many different industry practices. Among other things, I've attacked lootboxes, addictive mechanics, and the ethics of game service operators many many times. Your fight is not with me, and impugning my integrity is uncalled for. Skepticism, yes, of course. But there's no need to accuse me of dishonesty.
To address specific points:
Cost to users per unit sold is not graphed, but it's in the dataset. I had to go find it per title by hand. This is pretty easy; the vast majority of the points on the graph are $50 or $60. Free to play games are excluded from the player price paid for MB because there isn't anywhere enough data out there on LTV for F2P titles. MMOs from when they were in the boom period were baselined at $150, which matches my personal industry experience from that time, and is actually probably a little under.
Cost to publishers per unit sold is simply not available. Units sold for most games aren't available, unfortunately.
Revenue per unit sold is even more confidential than dev costs, and I don't have any figures at all for it. It's very rare for it to come out except in segments that use ARPU, and even then ARPU tends to be baselined on per day not lifetime revenue. Investor calls tend to only announce total revenues, not broken out by title.
"Value of game to a consumer" is utterly subjective, of course. I make no pretense at assessing that. You might highly value Mortal Kombat games; I don't tend to like them. I think Edith Finch was one of the best experiences of the year, you might prefer Zelda or Odyssey or Nidhogg 2. At best we could try for something like hours played, but that's an even less available figure.
So yeah... bytes are a very rough proxy, and far from an ideal one. They are just a (perhaps poor) stand-in for work product. They go some ways towards representing the amount of work that goes into a game these days, which unfortunately many game players seem to be unaware of. They are not a perfect proxy by any means. But part of the point in using them is that we can then look at a highly systemic game with low art costs that still did very well (like FTL, around 180MB, as you mention) and see that it is an outlier on the curve.
It is, unfortunately, incorrect that presence of absence of bytes has no value to a consumer. It's certainly not any sort of direct relationship, and it's not universal. But I think it's pretty fair to say that for a content-driven game, bytes are a pretty noticeable value. Graphics; levels and maps; characters and character choices; number of enemies... these are all things that players can and do use to measure the relative merit of content-driven games and we have all seen it. Not a universal, but common enough that it would be foolhardy to completely dismiss it. Byte expenditure is in many cases basically a marketing tool, as it contributes to art direction in particular. Cool visuals get you noticed. Do cool visuals always correlate to high byte count? No, and that would actually be a trend we would (and do) see as developers seek out cheaper yet well art-directed visuals.
As far as EA or other AAA shops. You can actually go look, they have in fact given percentage figures for incremental digital revenue versus retail, in some investor calls. The key thing is mostly that digital revenue is rising year on year. There is absolutely a debate to be had about "what margin should a publisher make" and "how much upselling is appropriate given that I already paid a retail cost." That's not what this article was about, though. I make no pretense to speak about profitability right now in this article. The point of it is the trendlines.