Bug Aesthetics

In my current day job (graphics programming for industrial simulators) I was recently writing some code to load volume textures. I made a mistake, and instead of seeing a rocky surface I got a pearlescent glowing road of colour:

Pretty Accidental
Pretty Accidental

When I flew around this shimmered something like a CD under a lamp. I took a screenshot because I liked the way it looked, and wanted to remember it. Sadly it still counted as a bug, so I had to fix it and move on.

But this made me realise I’ve done this a few times in the years I’ve been working in games and graphics programming. I dug around on my hard drive and found a few more screenshots from past projects. Quite often a bug or glitch can become accidental artwork, maybe even more interesting than the original intention.

Once I wrote a shader to make surfaces appear wet when it was raining, and then accidentally applied it to the sky:

Wet Sky

When the graphics card in my home gaming PC broke while I was playing TrackMania major glitches ensued:

Loading Screen
Race Screen
Main Menu
Main Menu

Formula One 2009 on the PSP. It took a while to find the missing pit crew. They were busy getting crazy under the track:

Pit Crew
Pit Crew

When image buffers weren’t get reset each frame an early build of Substream produced this:


In my modelling tool Heditor, the smoothing algorithm had an intermittent problem that created spikes instead. This took a while to fix but it looked kinda nice in it’s own way:

Smoothing Fail
Smoothing Fail


The Trouble with Doubling It

“If you want to know how long it will take to make your game, work out how long you think it will take, then double it”. This is a common rule of thumb not just in indie game development, but AAA game and software development too. The multiplier might vary with the experience of a company or team, but the general idea is widely understood. I took on board this sage advice when I started working on Substream. I calculated that my savings would last me about fourteen months, so I set to work on my dream game, knowing I should aim to finish it in seven. Three years on I have a full time job again, and although I still make determined progress on it by moonlight, things could have gone better. Believing in this rule of thumb is a psychologically complex problem, and I think it actually made things worse in my case. Here’s what tended to happen;

  • I designed a game to make in time X. I believe in the rule so I expect to finish it in time 2X.
  • I have lots of ideas as I go. It’s stuff that I’m sure would make the game better. I know I have 2X time, so when I ask myself whether I can fit this new thing in, it seems like I can.
  • Slowly but steadily, the extra spare time I’d given myself starts to fill up.
  • Surprise! At the end of time 2X, I’ve started integrating X more features into the project, so now I expect those to take another 2X.

It’s pretty easy to see the logical hole in this. I shouldn’t have added new features and stuck to the initial plan. That’s true, but when you see a way to make the game you’re working on better it’s hard to ignore even when you don’t think you have the time. When you have given yourself ‘spare’ time the justification just falls into your lap. Because isn’t this what that spare time is for? Unforeseen work? I’m not sure there’s an equally catchy rule of thumb or single bullet to prevent this way of thinking. The first step is to admit I am going to have lots of ideas and there will be a strong urge to fill my time. From this it’s clear that allowing myself 4X time is not going to improve things.

One  approach I want to try out in future is to have all the ideas I can, write them down, and completely fill up my available time budget with brilliant ideas. Then I will heartlessly cut out half of them in such a way that the game still works. The remaining work goes up front.

A Nicely Filled Schedule
A Nicely Filled Schedule

Now I have a two part plan. My time is already full, but if the rule holds true then I have still made a game in an acceptable amount of time. I’d end up with a finished lower quality game, rather than half a higher quality game. That’s a much better point at which to decide whether finding some extra time to add additional features is worthwhile. That’s the decision that’s hard to say no to when I think I have unallocated time.


Welcome to Video Games

I’ve seen people talking about a Financial Times article on Twitter over the last few days. It’s a unique insight; a view on game design and game culture from someone very new to games. I recommend reading it. The link is below, and the rest of this blog post gives my reaction.

Game Theory – What happens when someone who had never played video games is chosen as a judge for some gaming awards?”

I was actually blown away by this, and I think this a major achievement.

I don’t worry a great deal about the “wider cultural acceptance” of videogames. I feel like the share of the population that are gamers is large enough and invested enough that they can feel confidence in it’s worth without outside approval. But yes, the outside view of games is very different to other popular mediums. Most people who don’t regularly read books, visit museums or watch movies still have the understanding that they can create a wide range of emotions, educations and experiences. Many people who never play games do not have that understanding.

So when a person so far from the traditional video game target-market, approaching games with that background and attitude can come away not just having gained some understanding, but some enjoyment and respect is huge.


Not convinced? Try to imagine if this piece had been written in the PS2 era. Remember when “indie” wasn’t a thing, and video games could only find a market through a handful of big publishers who played it safe. When game designers repeatedly failed to pitch any ideas that strayed too far from stagnating genres. In the late nineties there was no platform to promote a game like Proteus, and no support for a game like Journey. Business models for games like this seemed impossible.

Surely the same writer’s reaction to a selection from the late nineties would have been very different. Whether or not you think there were games of artistic merit then, I think that someone with this attitude and background would not have reacted nearly as positively to the best offerings of those times. I can think of a handful of major reasons for this and some can be spotted in the article itself. It’s not my goal to go into those here, because my point is simply that these are being addressed. That’s great.

When I said “I think this a major achievement”, that doesn’t mean that I think that this article means “wider cultural acceptance” acheived. But the surface is scratched. The major achievement is that there are now games which someone who has these stereotypical negative predispositions can approach and come away from with respect and some positive insights. Obviously we still need a lot more games which do this, and do this better. Maybe we need more platforms like GameCity to set this kind of connection up. Simpler and more experimental games have been around for a while now; sometimes it’s good to remember the background they emerged from every now and then.


Who Decides “Is It a Game?”

Dear Esther is a recently released indie “walk-em-up” which originated as a free Half-Life 2 mod. It focuses on story and exploration with no combat. I’ve seen a few online discussions asking whether this constitutes a game. There are some reviews on gaming websites which immediately suggest that it isn’t. I’ve seen these kind of arguments about projects such as Minecraft and Sleep is Death in the past as well.

I’m not about to quote a definition for “game” here.

In the UK in 1999 if you said you were “texting your friend” you would have been some kind of idiot. Your English teacher would have cracked a wooden ruler over your knuckles and told you that you were “composing a text to” because “text” is a noun. A few years later, “text” was added to the Oxford English dictionary as a verb because so many people were using “texting” this way that it had taken on a new meaning in the English langauge. Dictionaries are there to reflect what people mean by words, not to lay down the law.

A review might be right to suggest that Dear Esther doesn’t fit a particular definition of game, but when you scroll down to the comments and see that it’s players “favourite game so far this year” and that they “would recommend the game” it’s the definition which is going to need updating. Gaming is growing and new genres are appearing. Not all games are going to be competitive or goal driven, but I’m sure the word “game” will get carried along with this movement. Whether something is a game or not will largely be down to the public.


Increasing the Awesome

I just finished Hexen II. I hadn’t played it before, I literally bought this game just because I loved Quake 1. But I have completed Quake many times and it holds no surprises for me now. Hexen II was developed with the Quake 1 engine, it has similar feel and style but is more RPG based. I thought I’d give it a go.

There was one particular puzzle I’d like to describe for anyone interested in game design…

  • There are nine buttons on the floor in a 3×3 square, and you have to press three of them in the right order to proceed.
  • There are three possible solutions, which are on a small unmarked plate on a random wall half an hour back in the game.
  • Each time you attempt to enter a solution, the ‘correct’ solution is selected randomly from these three possibilities.
  • This means it’s possible to try all three solutions in turn and not succeed.
  • If you press the buttons in the wrong order, you are teleported back through the level and have to walk for a minute to reach the buttons to retry.
  • Also when you are teleported back, the helpful message “Nothing happened…” appears on the screen.
  • After you press each button in, you have to stand nearby and wait for about two seconds for the button to rise before stepping onto the next one in the sequence, or this too counts as a failed attempt with the same message.

The only reason I know all of these conditions is because I read two different walkthroughs – neither of the authors of these FAQs were aware of the complete set of rules and mentioned that they were still confused and their descriptions may not be accurate. By figuring out some of these rules and with sheer determination, they had fumbled past this puzzle in order to continue.

Hexen II

This was the worst puzzle in the game but there were a few that came close. In one level I needed to find a spade, and although there was one was in plain sight in the middle of the level, I couldn’t pick it up because it was just a prop – the real one was hidden in a closet. I was also often frustrated by other flaws, such as doorways in the corners of rooms which were completely unlit, making the black doorway indistinguishable from the surrounding black walls. And I discovered that if I stood behind the final boss, he couldn’t see me so just stood there while I shot him in the back.

What did the critics make of Hexen II? The first three reviews I find on Google are MobyGames giving 82%, Gamespot giving 73% and CVG 94%. Fact is, in 1997 flaws like this were acceptable.

Some people say “games aren’t as good as they used to be”. I disagree!… Compare with Portal 2. OK, it’s not an RPG but it’s a first person shooter that tells a story, with lots of puzzles and some combat. And keep in mind that it also had similar review scores. But drop the puzzle I’ve described above into Portal 2 and you’d halve those scores, because design like that is not acceptable today.

I can be nostalgic about old games, but not especially so. I’m not too bothered about riding the cutting edge either. My favourite games are scattered liberally all over the last twenty years. My favourite Mario game is Super Mario Bros 3, but this is largely because I first experienced when it was released. I feel that it’s the best Mario game for it’s time, but that doesn’t make it the best of all time. If someone was new to gaming I’d be more likely to point them to Super Mario Galaxy, and I think they’d find this to be a more enjoyable game.

Graphics, sounds, story telling, control, menus, feedback, and presentation have all improved. People are better at designing games now. Putting light sources over doorways so the player can see them is just one little trick designers have learnt. But it goes further than that – they put the doorways in considered positions, or create architecture that leads the eye towards points of interest, so the player naturally finds their way through a level. They put play testers in front of puzzles and watch how they react, not just to check that they understand what they should be doing, but to see whether the solving the puzzle has made the player feel clever or educated them about the game’s universe.

Portal 2

If you look at some sub-par modern games and compare them with those retro games you cherish then things might look bleak today. There are modern design directions I dislike; achievements for doing otherwise illogical tasks, meaningless patronising reward credits, and I’m sick of customising avatars before I start every other game. But I think the average games of the times are better than the average games of past years. New games are designed with the experience of previous iterations. Each generation refines every aspect of game design. The best is yet to come.