Shrink To Success

Shortly after the success of Portal, a movement for shorter games began to form. It hasn’t gained much momentum in the mainstream, but game designers are beginning to recognize the advantages of “short-form” games, and I predict that producers and publishers will join the chorus within the next few years.

When I say “short-form” games, I’m speaking comparatively. A short-form game, for purposes of this discussion, is one which is significantly shorter than a mainstream AAA title; specifically, a game of roughly 2-4 hours in length, regardless of genre. The establishment of specific duration criteria implies that the game is at least somewhat driven by narrative, if not largely so; for example, it would be difficult — if not impossible — to quantify the duration of Tetris, so such non-narrative games are omitted from this discussion.

Short-form games can be uniquely compelling in ways their longer cousins can’t. For starters, their shorter duration necessarily eliminates “filler” gameplay, resulting in a more-or-less uninterrupted state of player education. This is compelling because, at their core, games are learning machines. The process of play is a cycle of learning, then applying:

learn-apply-cycle

When new concepts are being learned — whether they’re mechanics, environments, characters, or plot situations — player engagement rises. When known concepts are subsequently applied, or “tested”, engagement may begin high, but quickly falls as the application of the concept becomes repetitive; this is “filler” gameplay. The education of new concepts introduces novelty into the game flow and helps maintain player interest over time, but filler — too-long periods of application without learning — breaks the game flow and ultimately bores players.

Short-form games don’t have room for filler; thus, short-form games need not break the flow of education. New concepts can be educated to the degree necessary to ensure mastery, then applied for just long enough to provide validation of the new skill (and no longer!) The learn-apply cycle is compressed, resulting in greater overall player engagement:

learn-apply-cycle-compressed

Short-form games also tend to be more focused, in terms of both gameplay and story. Less overall content means room for fewer mechanics and fewer plot points, so short-form designers have to make the most of what they have. Portal, for example, has very few mechanics:

  • Fire blue portal
  • Fire orange portal
  • Pick up (and drop) objects (e.g. Weighted Companion Cube)
  • Turrets
  • Crushing pistons
  • Jump pads
  • Bouncing energy balls (with matching conduits)

By contrast, Grand Theft Auto IV:

  • Driving (cars)
  • Driving (motorcycles)
  • Helicopter piloting
  • Melee combat
  • Gun combat
  • Cover system
  • Cell phone
  • Bowling
  • Darts
  • Pool
  • Drunk-driving (and drunk-walking)
  • “Wanted” system
  • Vigilante missions
  • Internet cafes
  • Clothing customization
  • Safehouse customization

That GTA4 has a longer list of mechanics than Portal is neither surprising nor disturbing. But try ticking off, for both lists, the mechanics which were implemented poorly, and a very different picture quickly emerges. A major strength of short-form games’ necessary focus is that while they contain fewer mechanics, those mechanics are generally of a more uniform, and higher, quality.

The same goes for game stories. The stories in Braid, Knytt Stories, and World of Goo are simple, digestible, and most importantly, thematically and ludo-narratively coherent. By contrast, the stories of games like Metal Gear Solid 4 are sprawling, often incomprehensible, and packed with useless information and a low proportion of memorable moments. Not that I support games uncritically aping film, but there’s a useful maxim in screenwriting that applies to writing scenes: “Get in late, get out early.” The idea is that by presenting only the most irreducible core of a scene you increase audience comprehension of that scene, and by extension its impact and memorability. Put another way: distracting the audience with irrelevant or redundant content not only makes that content suck; it also drags down the perception of the “good stuff”.

So far, it all boils down to focus: short-form games are necessarily more focused than long-form ones, and therefore less likely to break the flow of player education or distract the player with meaningless content, leading to an ultimately more engaging experience. But there are several financial benefits to short-form games as well, and these are arguably more likely to make allies of producers and publishers.

First, the average gamer is 35 years old. That means he or she is likely to be employed full-time and to have a family, or at least a spouse. This is not a person with a lot of free time. Films are an enduring entertainment medium because they run about two hours, which isn’t difficult at all to fit into an otherwise busy schedule. 60-hour gaming extravaganzas are a whole different story, but a quality 2-4 hour game can be completed in a sitting or two and not feel like an endless slog. A ubiquity of 2-4 hour games would place games firmly in the same “impulse buy” space as movies and music, for the simple reason that consumers would no longer see games as a significantly greater time investment than those mediums.

However, even if such a ubiquity came to pass, games are still seen as a financial investment, due to the unconscionable $59.99 standard price point. The most common argument given by publishers at the beginning of the 360/PS3 generation for the price increase was that “next-generation” games were more difficult and, critically, more expensive to develop. While it is true that game development budgets increased significantly from the PS2/Xbox generation to the 360/PS3 generation, they are still, with the exception of a very few outliers, far below the average budget of a feature film. The key difference, as I have argued previously, is the difference in audience size. $10 DVDs sell to tens of millions of consumers and make a handy profit on $100 million dollar films. If a game sold to tens of millions of consumers, it wouldn’t need to be priced at $60 to make a profit on its measly (by comparison) $20 million development budget. And if you’re wondering where we find those extra customers, all you have to do is lower your price point.

But for the sake of argument, let’s indulge these publishers’ assertion that higher prices are necessitated specifically by higher development costs. Short-form games have dramatically lower development costs than long-form ones; notable short-form indie titles like Braid and World of Goo were done in their entirety for under $200,000. Such lower development costs should make reduced game pricing a non-issue, and as Valve’s data shows, reduced game pricing is highly likely to result in dramatic increases in net revenue… and that’s before we even factor in the short-form game’s superior focus, consistency, and ludo-narrative coherence, and the subsequent word-of-mouth and goodwill boost it’s all but guaranteed to receive!

It’s no coincidence that games whose duration is nearer that of films are likely to prove more compelling in many respects than 60-hour epics. Short-form games necessarily hold sacred the all-important flow of player education, keeping engagement high by avoiding filler and redundancy. They are more tightly focused, presenting a better-integrated set of mechanics and superior ludo-narrative coherence. They are cheaper to produce, which means they can be sold at a price point that supports impulse purchases, and good data suggests that price-point will actually increase net revenue. They fit sensibly into the average gamer’s schedule, all but eliminating the negative perception that games must be a significant time investment. And perhaps most importantly, they are small enough to be memorable in their entirety, rather than recalled in disparate pieces tainted by a plurality of poor experiences.

Presented with such a win-win package, why would producers and publishers not jump on board?

(This article was originally published at Game Design Aspect of the Month.)

Outsourcing the Boring Bits?

I was just reading this interview on Gamasutra with LOVE creator Eskil Steenberg (yeah, I’m a few days behind the curve right now), and about halfway through the interview he dropped this bomb:

Outsourcing, to me, is very stupid because you don’t get to keep the talent in the building. If what you are doing is so boring that any sweatshop can do it, you should spend time developing tools that do the job for you.

My initial response was along the lines of, “Ooh, sick burn! Devs totally make boring games!”

Then I remembered that I’m a dev, and I’ve worked on projects where we’ve outsourced work, and I’ve been proud of those projects. Ok, so maybe outsourcing isn’t automatically a stupid thing to do.

But now that Eskil said that, I find myself thinking a bit more deeply about outsourcing and its now-prominent role in game development. I’m continually surprised at just how much outsourcing goes on, and increasingly, what gets outsourced. I had always thought those third-party contractors were used as the game development equivalent of in-betweeners in traditional animation, but more and more I hear about key scenes and characters, concept art, even entire chunks of the design being outsourced.

I wonder, how far down this rabbit hole can we go before it becomes unsustainable? I don’t think we’re there yet — great games are still releasing on a fairly regular basis, and the last two years have been so fruitful as to recall the boom of 1998 — but where do we cross that threshold to where we’ve handed off so much of the game that we’re not even in control of our own project any more?

As a bit of a corollary, part of Eskil’s point is that outsourcing can potentially be replaced by in-house procedural creation tools. As I’ve watched the content requirements for games absolutely soar when compared to the PS2/Xbox generation, it occurs to me that we’re trying to solve the problem of not having enough manpower to fill our content needs, by simply moving the responsibility to other people. But come time for the PS4/Xbox 720 or whatever the hell they’re going to be called, are our outsourcing partners going to run out of manpower, too? Is a future of procedural content generation actually inevitable?

I wonder if in 2020 our outsourcing budgets will be paying, not for man-hours, but for time on massive cloud-computing farms that’ll be algorithmically building our content from scratch.

Utah Benefits From 4-Day Workweek

Last week I criticized comments by Epic Games president Mike Capps and Gears of War 2 producer Rod Fergusson that crunch time “is necessary” and that a “mere” 40-hour workweek was “kind of absurd”:

Mike Capps, head of Epic, and a former member of the board of directors of the International Game Developers Association, during the IGDA Leadership Forum in late 08, spoke at a panel entitled Studio Heads on the Hot Seat, in which, among other things, he claimed that working 60+ hours was expected at Epic, that they purposefully hired people they anticipated would work those kinds of hours, that this had nothing to do with exploitation of talent by management but was instead a part of “corporate culture,” and implied that the idea that people would work a mere 40 hours was kind of absurd.

[Rod Fergusson:] “I am a believer that if you’re going to make a great game, and there is that caveat, I believe that crunch is necessary. I believe it’s important because it means your ambition is greater than what you scheduled out. Going in with that idea that crunch is necessary means you can plan for it. It shouldn’t be a surprise. Crunch should be driven by the ambition of the team, and not the inaccuracy of the schedule.”

Yesterday, NPR posted an article about the Utah state government’s widely-reported move to a four-day workweek. In short, government employees now work four ten-hour days, instead of five eight-hour ones. The idea was to save energy and thereby cut costs, but it turns out that the real (and apparently unexpected) benefit of the change is increased employee productivity and satisfaction:

A recent assessment of the program by state planners found the expected energy cost savings haven’t materialized, but there have been unexpected boosts to productivity and worker satisfaction. [...]

Mike Hansen, strategic planning manager in the Governor’s Office of Planning and Budget, says one of the more surprising effects of this workday change is that employees are now taking significantly less leave.

That’s increased productivity — that’s employees behind their desk more this year than the last two years, to the tune of 9 percent” less time off, Hansen says. Paid overtime is also down.

While it’s probably true that working in government is not the same as working in game development, this result of Utah’s experiment appears to support the importance of a healthy work-life balance, which directly contradicts the attitude that’s recently been advanced by Epic.

Is it possible that the game industry could learn a thing or two from government?

Epic Crunch Is Epic

There have been a lot of games that fed into my desire to be a game designer, but at the top of the heap sits Unreal Tournament 2004. A great many of my opinions about shooter design, level design, and tools design are strongly inspired by that game and its excellent engine and toolset, with which I became intimately familiar during my work on Gem Feeder and various custom maps. I came to hold the opinion that Epic, the game’s developer, was the pinnacle of development studios doing the kinds of games I wanted to work on. Simply put, I made it my career goal to become a game designer for that company.

I applied, and was shot down — probably due to having no industry experience at that time, so it was no big surprise — but I kept that goal alive. Persistence is key, or so they say.

I got into the industry via another studio, planning to work my way up to my goal. A few years passed, and the first tech demos for Unreal Engine 3 started to surface. The new tech got me really excited, reaffirming for me Epic’s supremacy as a developer. Then Gears of War launched, and I loved it. That game showed that Unreal Tournament wasn’t just a fluke, and that the UE3 tech they’d been showing off was the real deal. I thought, “For sure this is the direction I want to go. I need to work on these games.”

Well, I’ve been in the industry for five years now, and I still don’t work for Epic. But now, I’m not so sure I want to. The biggest part of that has to do with me being incredibly happy with the team I’m with now, the project we’re working on, and my role in it all. I really couldn’t want for much better, and I’m surprised and delighted to have stumbled into such a great situation so early in my career. But there’s another side to it, and that’s some of the comments that have come out recently from Epic president Mike Capps and Gears of War 2 producer Rod Fergusson.

First there was this, via Greg Costikyan’s blog Play This Thing:

Mike Capps, head of Epic, and a former member of the board of directors of the International Game Developers Association, during the IGDA Leadership Forum in late 08, spoke at a panel entitled Studio Heads on the Hot Seat, in which, among other things, he claimed that working 60+ hours was expected at Epic, that they purposefully hired people they anticipated would work those kinds of hours, that this had nothing to do with exploitation of talent by management but was instead a part of “corporate culture,” and implied that the idea that people would work a mere 40 hours was kind of absurd.

Capps was backed up by Gears producer Rod Fergusson at last month’s GDC (via GamesIndustry.biz):

“I am a believer that if you’re going to make a great game, and there is that caveat, I believe that crunch is necessary. I believe it’s important because it means your ambition is greater than what you scheduled out. Going in with that idea that crunch is necessary means you can plan for it. It shouldn’t be a surprise. Crunch should be driven by the ambition of the team, and not the inaccuracy of the schedule.”

I do not believe in crunch. It is not a necessary evil, and it’s certainly not required to make a great game. Crunch is the result of a failure of schedule, or a failure of discipline. Period.

The dominance of the 40-hour work-week is not an accident. It’s backed up by reams of empirical evidence dating back to the beginning of the 20th century, evidence that has continued to support the 40-hour work-week despite the massive changes in our society over the course of 100 years. It is well-known and well-proven that productivity drops off dramatically beyond 40 hours, especially when overtime hours are worked for many weeks in succession. The IGDA’s Quality of Life initiative compiles much of this data in their white paper, articles, and presentations.

Productivity aside, 60-hour work-weeks wreak havoc on family and personal lives. Good luck having any hobbies outside of work (unrelated hobbies being a key advantage for game designers, incidentally). Good luck starting, or keeping, any kind of meaningful relationship. Good luck raising your fucking children.

Consistent overtime turns humans into drones, which is the kiss of a death for any creative endeavor. You would think one of the world’s foremost game developers would understand that, would feel some sense of professional and social responsibility to use their considerable influence to encourage sane working hours and healthy work-life balance in an industry that so desperately needs these things.

Epic need only advocate for good quality-of-life, or at the very least not advocate for crunch. But instead, two of Epic’s top people are trying to sell us all on the idea that crunch is good. That’s a tough pill to swallow, especially when Mike Capps made his comments at the Leadership Forum while seated on the IGDA Board of Directors.

I’ve held a lot — a lot — of respect for Epic over the last ten years. I’ve consistently held up their tools as the right way to make games, and pushed the teams I’ve worked with — both amateur and professional — to learn from what we see there. But Epic should hold their work-life balance to the same standards as their technology. It would be good for their employees, good for their image, and good for the industry at-large.

Our responsibility to society increases with success. This applies to individuals and corporations, in the games industry and beyond it. With great success comes great power, and with great power, as they say, comes great responsibility. And I strongly believe that a responsible game studio is one that rejects crunch, embraces work-life balance, and still produces great games.

The Myth About Game Pricing

We have yet more evidence that video games are too expensive for their own good, in the form of Gabe Newell’s 2009 DICE Summit keynote address. Rock, Paper, Shotgun has the story, and some truly damning statistics:

The recent Left 4 Dead sale lead to a 3000% increase over the previous numbers. That is, more than in the weekend it was released. Plus, another 1600% in new customers to Steam.

The holiday sales lead to interesting numbers. A 10% reduction lead to 35% increase in amount of money which came in (i.e. Not just sales). 25% lead to a 245% increase. 50% lead to 320% increase. And 75% lead to 1470%.

(The Left 4 Dead sale dropped the normally-$49.99 game to $29.99 for a week, if my memory serves me correctly.)

Off-the-cuff intuition would seem to suggest that a price drop would result in more units moved, but not necessarily more overall revenue, due to the reduced income-per-unit. However, Valve saw significant revenue increases right out of the gate at 10%, and the jumps kept getting larger the lower they cut prices!

Obviously there’s a point of diminishing returns, hovering somewhere around market saturation. Specifically, there’s a finite number of people who are even potential customers, based solely on their personal interests. That is, even if your game was free, that doesn’t mean every person on the planet is going to want it. So once you’re moving enough units to approach market saturation, further prices drops are going to stop increasing sales volume and start reducing revenue.

But until then, the evidence seems clear: the more affordable the game is, the more likely it is people will buy it. So much so, that the increase in sales volume more than makes up for the decrease in income-per-unit.

I’ve said this before, and I remain an advocate of lowering the price of games. The argument that game pricing has gone up to keep up with the increased cost of development is, in my opinion, arguing for the wrong solution. As Valve’s numbers clearly demonstrate, lowering the price increases the revenue dramatically, making it far easier to offset development costs than the standard price jump from $49.99 to $59.99 back at the start of this console generation.

Game pricing also has implications for piracy, as Cliff Harris noted (and I commented on) last year:

A LOT of people cited the cost of games as a major reason for pirating… People talked a lot about impulse buying games if they were much cheaper.

Are there really any arguments left for maintaining the $59.99 price point?