The cost of games

https://www.raphkoster.com/2018/01/17/the-cost-of-games/

by napsterbr • 9 years ago

137 76 9 years ago

megaman22• 9 years ago

Play time in my Steam library is overwhelmingly tilted towards just a few titles. Overwhelmingly, these are older, sandboxy games that have had strong modding scenes. The number one game in my library is Medieval Total War 2. I believe I paid $10, on sale, for it plus all of its expansions. After Steam, and Sega took their cuts, Creative Assembly got a pittance, and I've gotten over a thousand hours of gameplay. The community mod makers that expanded that game massively (those that didn't end up hired by CA, anyway) never made a red cent for their labors of love. I love that game, and it's a tremendous value for me.

Much as I dislike it, trickle out DLC and in-app purchases are one way to try to extract more value out of that massive investment. But it's a tough proposition, going against everything that already exists. I could never buy another game, and have more gameplay available to me than I could ever exhaust in multiple lifetimes.

People will always make games, because they love doing it. But it's not really rational; you've got to be good and catch lightning in a bottle

earenndil6• 9 years ago

There are two forecasts at the end based on exponential growth. The thing to remember about exponential growth is that it is inherently unsustainable. Consider Moore's law. I once read part of a book from 2005-ish about the singularity whose basic premise was that by ~2050 or so we'd have enough computing power to stuff orders of magnitude of orders of magnitude more power into a now-sized CPU. And this was in a completely serious tone, mind you. Now we see the slow demise of Moore's law over the past 3-5 years and it seems ridiculous that we could have ever believed something like that just a decade ago!

The state of advancement in the fields of human society, technology, and science, at their current rates, is such that around 2500 all of human society will be overturned every week, day, hell, minute or so. Which obviously is nonsensical.

Note also, however, that forecast #2's best-fit line doesn't seem to fit the points very well...

singularity2001• 9 years ago

Exponential growth can be sustained for millions of years, see life on earth. Moore's law lives on in GPUs and TPUs.

w0utert• 9 years ago

Except that it doesn't. There are no forms of life on earth that have been growing exponentially for millions of years. Sooner or later growth and expansion always hits its limits and populations stabilize or decrease.

Sacho• 9 years ago

But this is the problem of exponential growth - you can sustain it without even noticing the detrimental effects, until suddenly you can't.

The resource usage of all the doubling time periods until the last one would have used up only 50% of the resources available(this could be millions of years, depending on the exponent), and then suddenly the last one would use up all the resources left.

This is repeatedly exemplified in Professor Allen Bartlett's talk "Arithmetic, Population and Energy"(video link at specific point: https://youtu.be/sI1C9DyIi_8?t=1337 transcript: http://www.albartlett.org/presentations/arithmetic_populatio...)

neel_k• 9 years ago

Total biomass actually peaked 540 million years ago, at the end of the Proterozoic era. See Figure 2 (on page 88) in Franck et al.: Causes and timing of future biosphere extinctions.

Link: https://www.biogeosciences.net/3/85/2006/bg-3-85-2006.pdf

earenndil6• 9 years ago

Not really. Moore's law refers to the number of transistors that can fit on a die, and GPUs/TPUs haven't increased that.

sdhgaiojfsa2• 9 years ago

What aspect of life on earth has been growing exponentially for millions of years? I'm not clear on what you mean here.

earenndil6• 9 years ago

see life on earth

Notice how life on earth is plateauing...

scrooched_moose• 9 years ago

The difference between exponential growth and sigmoidal growth [1] is extremely difficult to tell in the beginning. I tend to believe everything is actually the latter, we just haven't gotten to the point they diverge yet.

1) https://en.wikipedia.org/wiki/Sigmoid_function

alkonaut2• 9 years ago

I was under the impression that Moore’s law was still alive and well.

Obviously game cost can’t sustain exponential growth if the market doesn’t grow exponentially (potential buyers times what they are able/willing to pay).

earenndil6• 9 years ago

Nope! Quantum tunneling has made it so that the size of transistors can only shrink so far before electrons start essentially teleporting one from the other. AFAIK we might be able to get to ~3nm lithography (TSMC is shooting for it but beyond by 2020 or 2023, IBM apparently already had 5 or 7nm (forget which) lithography but hasn't published it yet), but beyond that the future is much less clear.

orbitingpluto2• 9 years ago

"If you want to preserve the games you love, you can help by not pirating,"

conflicts with:

Did Nintendo actually sell us a Mario ROM on the Virtual Console?

https://www.technobuffalo.com/2017/01/22/nintendo-virtual-co...

Or if you go onto a retro gaming site like GOG you'll see that many of the "extras" are from pirated sources.

galapago• 9 years ago

Or if you go onto a retro gaming site like GOG you'll see that many of the "extras" are from pirated sources.

Can you give one or two examples of this?

Springtime• 9 years ago

I'd be interested as well. Certainly being DRM-free GOG.com's content typically ends up on torrent sites, but I haven't heard of it the other way around—though there are of course various examples of pirated content appearing in officially published downloads elsewhere I've read about and even experienced first-hand. One that comes to mind was 90s hip-hop act De La Soul generously releasing their discography for free during Valentines Day a few years back to promote a new record, which from some ID3 tags mentioned the source were pirated copies.

You have to wonder when an artist finds it easier to copy the albums from such sources than from the collection of their own.

orbitingpluto2• 9 years ago

I didn't devote much time, but about half of the 10 manuals I looked at quickly are of horrible, almost unreadable, photocopies. One thing I did find is that some games, like Arcanum, use "warez" copy protection cracks.

https://www.gog.com/forum/general_archive/gog_arcanum_releas...

PhasmaFelis• 9 years ago

That's not really relevant to his point, though. I'm sure it's quicker to get a ROM dump of Super Mario by downloading it than by digging it out of Nintendo's archives or setting up their own cartridge ripper. That doesn't mean they couldn't or wouldn't have done one of those things otherwise.

bitwize• 9 years ago

So, that's standard practice in the industry. Download your own stuff from a pirate site, then sue the shit out of the pirate site owner.

freeone3000• 9 years ago

So what if they did? They own the rights to the Mario ROM.

krapp• 9 years ago

If Nintendo lost or threw away their own copies, the software would no longer exist for Nintendo to use, apart from pirated ROMs. The implication is that software piracy is sometimes the only viable means of preserving software in the long term.

timavr• 9 years ago

Just looking on Bungie LinkedIn, they employ 800 employees. Industry standard burn rate per month is 10k. So around 96m a year minimum. They will have tonnes of outsourcing expenses, testing etc etc.

So Destiny 2 on 3 year cycle at minimum 300m on dev costs alone, most likely higher.

aaron-lebo3• 9 years ago

The worst part is it's a beautiful but poorly-made game. Even their biggest fans hate what they've done.

https://www.reddit.com/r/Games/comments/7raulk/bungie_has_ab...

https://www.pcgamesn.com/destiny-2/destiny-2-reboot-jason-sc...

What good is employing 800 people and spending 100 million a year to develop a bad final product? If you want to know where the tech industry can be attacked, it's in that massive inefficiency.

always_good• 9 years ago

If you want to know where the tech industry can be attacked, it's in that massive inefficiency.

How is this different from any large, ambitious project? It's also essentially an art project where you have to make content people enjoy, so engineering resources won't save the day like they can in tech. And you can end up in a place where your lead visionary ended up in a place where they just couldn't deliver on it or it just wasn't the necessary vision, so you have to replace them.

Seems weird to ascribe all this to "tech". Turns out, things are hard. To me, Destiny 2's failure has more in parallel with the fact that most people never finish developing a game nor writing a book.

indubitable• 9 years ago

Price per MB is a questionable metric that I think the author made insufficient effort to justify. The PS3, for instance, sent game sizes skyrocketing in an effort to try to demonstrate the perks of blu-ray. A number of games included all audio localizations on each disc in lossless formats. That likely saved studios money since localized distribution could all come from the same production line, but it sent the $/megabyte way down which this author implies means studios are earning less. It's a nonsensical datum without some justification and that justification was not provided in this article.

I also think that that datum leads to questionable conclusions. For instance you're paying vastly more for AAA games today than you were in the past. The author is right that the upfront cost is certainly much lower, but we've changed from a model where you're paying for a fully fleshed out game to one where you pay for a stripped down product that is intended to be sold piecemeal. In many cases users are paying to unlock content that is literally already on the disc they purchased. That first $60 you pay is simply your "starter pack." And there are additional 'meta costs' as well. Instead of simply putting the game in and running it you're instead obligated to pay an annual fee to use your own internet connection and then exposed to various advertisements on your way to launching the game and games themselves increasingly even feature more advertisements on their launch screens. These advertisements, in effect, increase the cost to the player if we assume they are effective.

Why not use profit margins? Total revenue / (total development + total marketing)? In either case, it doesn't really matter. Games are like music in that they are self sustaining since people will continue develop them even absent a profit motive. This is even more true today thanks to platforms like Steam removing the necessity of publishers from the entire industry. Seeing the end of 'fee-to-pay' or AAA games would bother me no more than seeing the end of Justin Biebers.

j_s• 9 years ago

Money is important, but is there a place here to discuss the 'time' cost of games? Games are a useful tool for relaxation and other purposes (and a comparatively inexpensive hobby) but in my case it is easy to get carried away! For my most serious addiction I have chosen to time-box my gaming by only paying for two months each update (every other year-ish).

It takes conscious effort and self discipline to change direction toward creation instead of consumption. The tiny bit of gamification here on HN has me completely hooked, but I cling to the notion that my participation is creating some value (however slight). I look forward to the day when maintaining CRUD web apps is properly gamified, so I can earn money playing games!

PS. There is another opportunity cost of games vs. alternatives which require strenuous physical exertion that cumulatively improve health and conditioning (exercise, sports, etc.).

saturdaysaint• 9 years ago

Is pirating still a huge issue? Granted, I'm not a broke college student any more, but with the ease of Steam and outrageous sales and giveaways, I'm amazed even broke kids would still bother.

falcolas10• 9 years ago

<rant>

This is a prime example of the practice of lying with numbers. There are two very basic "cost" graphs which are notably missing from this narrative, and the addition of "bytes" as a major factor in deciding the value of a game to players.

The two missing graphs are: Cost per unit sold, and revenue per unit sold. As basic as these are to any good cost/benefit analysis, their absence indicates that they do not agree with the narrative that the price of games needs to go up; that companies need to milk more money out of their customers.

As for bytes, the presence or absence of bytes has no practical impact on the value of a game to a consumer. If it did, Duke Nukem Forever would be a game worthy of game of the century, whereas Faster Than Light would be forever consigned to the scrapheap.

Perhaps the most telling in this narrative is EA's own comment to investors that shutting down the sale of lootboxes in Battlefront would not result in them losing money. This indicates that retail sales alone were sufficient to offset the cost of development, licensing, and marketing.

</rant>

I also don't condone piracy, but to imply that we should feel sorry for these companies who are barely scraping by is blatantly dishonest. It is especially dishonest when you consider that the bigger publishers (the ones making the outlier games on all of those cost charts) have annual profits in the billions while (oops, rant didn't end) they prey on those susceptible to gambling addictions.

AndrewOMartin• 9 years ago

Jim Sterling, is that you?

To help avoid accusations of not adding anything to the discussion; Jim Sterling is a video games essayist who publishes well researched and presented polemic on the games industry on YouTube.

falcolas10• 9 years ago

No, but I'm both flattered and bothered by the comparison. :D

Jim Sterling provides a viewpoint not commonly shared even among some of the more critical video game reviewers (such as that even cosmetic lootboxes are bad for the consumer), and it's provided a few insights I hadn't thought of previously. And yes, the EA quote is also one of his favorites - it is very telling.

He has created hell of a persona - one hell of an offensive persona - but it hides a lot of really useful insight and sly humor.

vvanders3• 9 years ago

I also don't condone piracy, but to imply that we should feel sorry for these companies who are barely scraping by is blatantly dishonest.

As someone who spent a fair part of my career in gamedev, yes the publishers are taking a large share but the companies that are building the games are barely scraping by.

Most gamedev companies are 2-3 milestone payments away from bankrupcy, which is partially why the publishers have such leverage. Implicitly condoning piracy by saying the publishers make a ton of money still ends up hurting the studios in the process.

falcolas10• 9 years ago

That's not what I said - that's not even what I implied. My not "feeling sorry" for publishers or game developers is not the same as condoning piracy implicitly or explicitly.

But since you brought it up, let's go there.

The primary effect of piracy is the loss of sales, which if big enough would naturally lead to the loss of funding for future projects, leading to a studio being closed down. However, very few studios that make big budget games are closing down because of good games failing to become profitable - they're closing down because of politics or they made a game that people don't like.

Even Bioware Montreal, one of the bigger studios attached to a "flop" this past year wasn't closed down because the game resulted in a loss for EA; "Mass Effect: Andromeda" was, per EA, still quite profitable. Instead, Bioware Montreal was closed due to situations largely caused by mis-management (at least, based on what information we have available to us). Most of the developers were even retained.

Smaller, self-published studios would, of course, feel the pinch of piracy more than larger publisher-backed studios. That said, they're also playing in a very crowded pond, and differentiating between a failure caused by piracy and a failure due to an over-saturated market are going to look ridiculously similar. It would take a crystal ball to tell the difference.

Finally, let's be frank: all game developers big and small are affected by piracy, so it's a mostly even playing field on which they are competing. A studio that fails and blames it on piracy is not looking hard enough.

Sorry, there are some rough words there, but I stand by them. I don't condone piracy, but I also accept that it's simply part of the world in which all game developers exist. Reality is not always nice.

lj32• 9 years ago

their absence indicates that they do not agree with the narrative that the price of games needs to go up

Their absence suggests to me that the writer of the piece didn't have access to that information. Wouldn't you need inside information to come up with those numbers?

falcolas10• 9 years ago

Well, here's the writer's own words on that subject:

Using industry contacts and a bunch of web research, I assembled a data set of over 250 games covering the last several decades.

I would make the reasonable assumption that revenue could be guessed at, if not explicitly uncovered, given the detailed data the author appears to have on the number of sales and costs. Many publishers and dev houses are public companies who release public information about their costs and revenues.

zitterbewegung2• 9 years ago

Instead of piracy just don’t buy the games. I usually wait a week or two after release to see if the game is even worth looking into.

falcolas10• 9 years ago

A good policy, IMO. And even if you know you'll buy the game at launch regardless of the reviews, there's no need to give the publishers your money (or the promise of your money) months in advance; a day or two should suffice.

bllguo• 9 years ago

So quick to claim malice and deception, when the likelier story is that cost/revenue per unit sold isn't public.

I would be hesitant before throwing out accusations and claims so irresponsibly. Have you done any analysis of game size vs. quality? Two outliers does not destroy a trend. There are arguments for it - only bigger teams with more resources will make large games. It would not surprise me if larger games tend to have higher playtime/better reviews. Consider the sea of small, trash games out there.

jblow2• 9 years ago

Not everything is a conspiracy. I know the author of this article and he is a straightforward person.

The game industry is a big place with many different kinds of people in it, all of whom have different motivations.

If you build a picture for yourself wherein everyone is That Guy At EA Who Put Loot Boxes Into Battlefront 2, then that picture will probably get in the way of your understanding what everyone is being said by everyone else in the industry who is not that one guy.

This article is targeted primarily at developers, not at gamers. It is an "oh crap what do we do" article, because believe me, that is a big problem and in general we do not know what to do.

This was not some "cost of games need to go up because blah blah" justification article ... it is a straightforward look at what has been happening and some musings about what we can do about it, with "cost of games will probably rise" thrown in as a 6-word aside near the end (at which time he also points out that nobody wants to do this, which is true).

You're reading the article you want to read, not the article that the author wrote.

lacker• 9 years ago

1. Thanks for making Braid that was great

2. I believe you that the author is a straightforward person and I don't think there's anything malicious here. Still, it's not a good dataset. It's 250 data points collected in an ad hoc way. A single decision like "do we include PUBG or not" can sway every statistic in here. So I don't think it's a good idea to make any conclusions based on this data.

falcolas10• 9 years ago

Motivations and character of the author matter little; the article which was written is pushing an agenda: a justification of (and if not a justification, a recommendation for) the use of what could be seen as questionable practices for reducing costs and getting more money out of their audiences.

What practices do I consider to be questionable?

"de-emphasize bytes in favor of other types of content"

This, and the rest of the suggestions reads as "don't build a hand-tailored experience, instead build more random permutations of generic content." There's even suggestions to find ways to monetize some of those permutations.

I'll admit it: I'm a gamer, not a professional developer. But these things matter to me. Good, profitable, games do exist - I am playing many of them. What bothers me as a gamer is that all of the suggestions being offered up here will never re-create one of these good games. The games being advocated for describe yet another vehicle for selling lootboxes or skins or weapons.

The whole article closes with the comment "I would love it if these graphs were wrong". Charitably, we can say that the graphs the author displayed are not wrong. However they provide, at a minimum, an incomplete picture. They invite readers to come to a very specific conclusion. That practice is what I'm criticizing with the phrase "lying with numbers" (credit for which goes to "Penn & Teller's Bullshit").

Noos• 9 years ago

You don't need any of this. I paid $70 for Final Fantasy 3 on the SNES in the 1990s. The average AAA PS4 game is 59.99 in 2018. Games are one of the few things that not only have not matched inflation, they've gone down despite the sheer amount of people needing to work on them going up.

falcolas10• 9 years ago

It's worth noting that the number of people who purchase games has far outstripped the losses due to inflation by several orders of magnitude.

wink2• 9 years ago

Anecdata - but I'd say in the SNES there was hardly a game that only provided 10h of linear game play, whereas that's quite a possibility nowadays. (More on PC probably, but I don't think all PS2/3/4 games are created equally). So maybe people wouldn't even buy it at 130$ (first google hit for converting buying power 1990 -> 2018)

watwut• 9 years ago

It is ok for companies to earn money. I am enough on the left to want better control over drug prices and to be huge fan of public school systems including college.

Getting angry over game companies wish to be profitable is too much on the left.

This is literally forum for entrepreneurs trying to earn money on whatever. Why should be game expanses expected to not seek profit here is incomprehensible to me.

Raph_Koster6• 9 years ago

Article author here.

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.

falcolas10• 9 years ago

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".

crispweed• 9 years ago

Maybe the solution is to reuse or share content a bit more? Seems wasteful, for example, that the huge worlds created for some of todays AAA open world games are used only for those games.

aaron-lebo3• 9 years ago

This is interesting coming from Koster who was the lead designer of some of the the most interesting MMOs ever made in Ultima Online and Star Wars Galaxies.

SWG had the misfortune of releasing in 2003, by 2005 EA was trying to compete with WoW and within a few years had nuked the game so much that they ended up canceling it. There is a great emulator out there for it, though the game at this point shows its age.

cthor• 9 years ago

Getting Over It: With Bennet Foddy breaks all these rules and was wildly successful.

The Witness sold at $40 and people bought it anyway.

Good games will sell. But you might not have a good game.

The market doesn't owe you anything.

Also what kind of bogus metric is "price per MB"?

sdhgaiojfsa2• 9 years ago

I don't think it's necessarily true that good games will sell. Some will, sure, but others may not. There are massive numbers of games that fly under the radar and never really catch on, and some of these are bound to be good. Like anything else, there's an element of luck involved. Simply being good is not enough.

Price per MB seems like an attempt to capture the amount of work that goes into a game, but I agree it's not a great metric. You can decrease price per megabyte just by shipping models and textures with higher resolution, even though they don't cost you anything more to create.

Raph_Koster6• 9 years ago

Getting Over It (which is awesome) is exactly the sort of deeply systemic game that the article advocates needs to become more common.

rpdillon• 9 years ago

I still play Morrowind regularly (it came out in 2003), and I sometimes wonder if gamedev would be more sustainable if we focused on increasing the half life of games. I think that approach is more viable with single player experiences, since multiplayer is more dependent on a critical mass of players.

fancyPantsZero• 9 years ago

Isn't this the opposite of the solution, though? If devs only made games that lasted 15 years, they would probably sell way fewer games in aggregate.

davewasthere• 9 years ago

Cost per bytes? Makes no sense at all.

However cost per hours played... That would be a metric worth exploring. Using some combination of how many people own it, how many hours played on average and cost of game... That would give you the 'value' of each game.

http://steamcharts.com/top

I think back to a game like Manic Miner which was less than 48kB, giving me hours more fun than say Don't Starve (still a good game), which cost me a few bucks on sale and I've played about half an hour... yet comes with a 600Meg install size.

The cost of games

Related Stories