Tuesday, October 18, 2016

The attempts to bring "AAA quality" games to mobile - and why they are failing

Last year, I wrote about my experiences in what I saw as a tech arms race in AAA PC/console games. To recap: I've gone from playing games because of their impressive use of technology to playing games because they're fun. I've slowly come to discover the obvious: The two are not corellated. I have grown increasingly frustrated by AAA's insistence on maximizing hardware use regardless of game's actual needs - an insistence that's largely responsible for out of control complexity and costs.


For reasons I mentioned in that article, I have no doubt the arms race will continue, despite rare insider speakouts against it. Already, both Microsoft and Sony have revealed their plan for more frequent hardware updates during a console lifecycle. In fear of being left behind by the competition, tech executives will mandate game teams implement features and content that pushes the new hardware to its limits. In response, game teams will complicate their technology base and overall development process even more, suffer more crunch, use more time and money to deliver buggier games, but with the usual diminishing visual improvements that we'll all be testing our vision with when staring at side by side pictures comparing how the game looked on earlier hardware.

Tech obsession on mobile platforms

Meanwhile, in the mobile games I've been working since, there is not really a tech arms race, despite the fact that phone and tablet hardware is getting closer to console quality and still advancing rapidly. That's because most of the successful mobile game companies are smart enough to see the race as a huge, expensive distraction and are instead focusing on far more important aspects of making and marketing games.

Still, there have been many attempts over the past few years to bring "console quality" games to phones and tablets. Some them have been by established AAA companies. Here's an EA executive revealing a very common attitude among AAA studios: that mobile games are not mature (they're not even "real" games) until they start participating in the tech arms race:
we’re going to start making games on Frostbite Go that will be featuring things that you’ve seen in the regular Frostbite code, like destruction and those types of things. We think we can do something dramatic there, with the mobile sector, and make real games on those machines.
Some mobile startup companies also choose to go through the very expensive investment of building a custom "console quality" engine, because they believe such advanced technology (and the people who can create it) will be the key differentiating factor for their game. For instance: 
Super Evil Megacorp sees its investment in proprietary engine tech as a beacon for attracting developers interested in building graphically-intense mobile game
It's not just traditional AAA companies and mobile startups seeing that approach promising. Kabam is now betting on the premise that games with "console quality" elements will do better. In an era where many other direct competitors are focusing on experimenting with finding fun concepts as fast and as wide as possible, Kabam seems to be taking a different course, going as far as to say "...we have cancelled games in development that are not AAA quality"

What does "AAA quality" mean, anyway?

If, like myself, you've been paying attention to gaming culture for a long time, you probably associate the term "AAA quality" with tons of hand-crafted content, advanced 3D engines that use the latest hardware to its fullest extent, complex/lifelike/realistic particle systems and animations, etc - the stuff that's typically so expensive, only companies who can hire 300 person teams can do them. Supporting this, Wikipedia tells us the term AAA is "...used for games with the highest development budgets and levels of promotion. A title considered to be AAA is therefore expected to be a high quality game..." This is what I, and the quotes above, are referring to when talking about "AAA" or "console-quality" games.

But outside games, "AAA quality" means the best quality product. If someone who didn't grow up with our definition read that Wikipedia page, they'd be inclined to believe that only the most expensive games are any good. That's a ridiculous notion.

We have all been so conditioned by this that we don't even apply the term AAA to games that are otherwise generally accepted to be very high quality. Minecraft? "Sorry, too pixely". Portal? "Maybe if it had more guns and lasted at least 8 hours, then it could be AAA". Spelunky? "What's wrong with you, it's not even 3D".

The fact that many of us consider "AAA" (as in "best quality") to be equivalent to technical excellency, great amount of content and complex 3D engines is a sad consequence of living in a bubble of our own making. I won't go too deep into it, because I covered this in my previous post. People who play mobile games don't live in the same bubble, or at least recognise that there are far more important elements to what makes a mobile game high quality.

When some companies say they are "bringing AAA quality games to mobile", they are primarily talking about bringing the technical excellence and amount of content AAA console games have, believing or hoping that such technical excellence will automatically translate into higher general quality of the game. But it is my position that the two are not correlated and, even worse, forcing a team to prioritize technology and content types found on console games is a bad idea on mobile.

"AAA-quality" games aren't doing that well on mobile

Judging by what we see so far on the app stores, it's fair to say that the "AAA quality games on mobile" approach is not working that well, despite support from the platform holders (Apple for example heavily showcases games that use the Metal API). EA had lots of console-like games (including Dead Space and Mass Effect: Infiltrator) removed from the app store because they weren't popular enough to warrant an update to work with the latest iOS version. No game comes to mind that has found any significant long-ish term success primarily because it pushes the hardware in ways other games don't. Just out of personal curiosity, for years I've occationally looked at the top 100 grossing games and asked myself the question, "How many of these games wouldn't be as popular if they weren't using advanced 3D technology that pushes the phone to its limits?". My answer has always been 0, though depending who you ask it could be argued that maybe 3-4 out of 100 games fit that criteria. Of course, being in the top 100 isn't the golden test of success for everything, but usually investing in AAA quality technology means bigger than average teams - for those teams, being out of the top 100 makes it much less likely to see a recoup on their investment.

Why are those attempts failing? There are some juveline arguments ("mobile players don't care about good games"). There's the classic arms race participant mindset that says "mobile tech isn't there yet". Better arguments note that such attempts almost always try to bring PC/console-style gameplay to mobile devices (with noteable exceptions, such as the Infinity Blade games). There are obvious arguments about why console-type gameplay is not really suited on mobile. I used such arguments when comparing two great games, one of which didn't really fit the mobile platform.


Oz: Broken Kingdom has impressive cutscenes during and between combats - but they interrupt gameplay flow and cause excessive loading times

I agree that attempts to bring "console quality/AAA" games usually come with baggage that's counter-productive on phones and tablets. Cutscenes (often unskippable) that completely mess up the flow when players are looking for a quick session. Excessive loading times (some times over 30 seconds!) and general sluggishness also discourage very common playing patterns. Massive content downloads on the launch that matters the most - the first one. Heavy content uses up more memory, making it very likely the game will need to do a full reload next time it is launched. Impressive "life-like" visuals sometimes make it too hard to distinguish foreground objects that really matter to gameplay.


The Collectables visuals were impressive, also made gameplay confusing

Different kinds of "best people"

I've written before about what I think can make small teams really stand out. I think those guidelines can help teams of any size, but are far more practical to implement in small teams, like the ones found in mobile games.

As described in that post, many companies talk about hiring "the best people", but without proper qualification, that talk can get in the way of hiring the ideal people for a given job. Qualifying "best people" as those having deep technical expertise (as companies attempting to bring "AAA quality" games to mobile are very likely to do) sets a very dangerous starting point.



When I was working on AAA games, I thought it was slightly weird for some developers to argue that players should not be able to skip cutscenes. "We worked hard on making it", "It's necessary to understand the story", "It's really cool", none of those seemed to me like a good enough reason to not allow a player to skip that content if that's what they really wanted to do. Still, it wasn't that big of a deal. If someone is sitting on a PC or console playing a long session game, they can reasonably expect to look at 2 minute cutscenes every now and then.

But seeing the same attitude on some supposedly "AAA-quality" mobile games is eye-opening. One of two things happened there: Either nobody on the team thought not allowing skipping of cutscenes is a bad player experience on mobile, or (more likely) the people who thought so were overruled. In either case, this is an example of how a tech-focused team can make patently bad decisions in the mobile space. Other decisions that are not as obviously bad can still creep up and significantly affect the player experience. For instance, not everyone is good at controlling a 3D camera. Merely forcing 3D with a rotateable camera because "it's cool" (like this team for example), is generally a bad thing because it adds complexity to the user experience (and to the development) for no good reason.

This is why I believe it is important when qualifying "good developers" to start from T-shaped individuals who have wide experience, are primarily driven by making good games and understand the platform, and then look at how they fullfil technical requirements - not the other way around.

When EA is building their mobile technology primarily around what has worked on consoles and delivers "destruction and those kinds of things", when Kabam is presenting their vision as "AAA-quality first" to potential new hires, when Super Evil Megacorp is using their engine to attract developers who primarily want to build "graphically-intense mobile games", they are all making it far more likely to end up with teams that are technology-driven than game-driven.

People who are obsessed with technology have different goals that people who are obsessed with the game they're making. They both may understand that what the player sees is extremely important, but they'll interpret that importance and seek to achieve a good visual experience in completely different ways. People who care about the game are more likely to seek game-specific solutions (often low-tech and easier to achieve), while also keeping in mind other limitations that affect player experience. People who care about the technology are more likely to seek techniques that are expensive (both on the hardware and on development), because they see that as their mission and a way to advance their craft.

With mobile teams being generally much smaller than AAA teams, having even one or two such tech-obsessed people can negatively affect the entire team. Suddenly, decisions that should be obvious turn into complex politics. Whether the game should load fast becomes a "matter that has to be balanced" with the amount of content. Whether the player is allowed to skip long cutscenes on a mobile phone suddenly needs long meetings to be decided. Concerns around the user experience are brushed aside, using prior console experience as a guideline ("it doesn't really matter if they wait to download a lot of content, it's just the first launch").

If you are thinking about betting your mobile future on "AAA quality" graphics, or working for a company that does, make sure the people you will be working with have the right priorities. And I wish you the best of luck in picking technology with the purpose of driving actual innovation in the mobile space, instead of just relying on the unfocused tech arms race that has worked so far on consoles.