GameCity5 - Day Three - Old games and new faces

This morning's Guardian breakfast posed some interesting food for thought: what are the games that marked the most significant milestones in gaming history. Clearly a contentious topic and I won't go in to the list here; I think that most people, when faced with the list, would come up with alternative games that they feel are more significant for different reasons.

It rings true though, that there are certain games which simply change something in your mind about how games work or the potential for games to work. In part of today I saw a group of graduates discussion their first commercial venture, QUBE. As the game is a first-person-puzzler (for want of a better term) it draws comparison to Portal, although there appears to be no desire to add plot or implied narrative to the puzzles.

These two things coming together reminded me what is so startling about Portal. In terms of gameplay it is an extremely clever way to frame a puzzle game. What is more remarkable is the way in which a rich narrative is told without any secondary characters (apart from the voice), cutscenes or story events. As a distinction from, say, Myst - which is an adventure game with puzzles inserted - Portal is a puzzle game with a rich, hidden narrative. It's like being presented with a wordsearch in which the shape of the lines form words across consecutive pages to read "HELP ME IAM HELD PRIS ONER".

All of which mental wanderings bring me to the second of Jonathan Blow's presentations which I attended about his new game The Witness. Loved by many for creative the 2D time-shifting platform-puzzler Braid, many were, perhaps, surprised to find that his new project is set in a 3D environment with first-person controls. Chris Hecker (Spy Party) was in the position, as friend of Jonathan, to ask the inevitable cheeky question: "So what do you say to the young developers we are always telling to avoid 3D and make 2D games?"

Jonathan laughs, caught out. "We always say," he tells the audience, "the worst mistake is to make a 2D indie game ... make a decent amount of money to fund future development and then try something ambitious and obscure in 3D that no-one wants to play and lose all your money ... I'd never recommend to anyone to do what I'm doing."

But the reason for this departure is a deep love of adventure games, or a least a deep love of what adventure games should be. Speaking on the death of adventure games, Jonathan says "other games evolved to be more playable - adventure games never figured out how to do that." Frustrated by a genre which mostly revolved around hoarding inventory items, randomly combining objects and failing to interact with scenery, Jonathan has set out to make a true adventure game which combines the joy of exploration with environment based puzzles which play by a fair set of rules.

He demonstrates a good set of the opening of the game and it is clear that the player is taught how to understand the puzzles in the game a little at a time, without any text displays or audio prompts. Simple line puzzles build into basic mazes and then evolve into complex mazes with pathing rules. Basic levers are identified early as well as active and inactive switches and through an elegant opening puzzles players are taught how power lines from nearby puzzle stations to environmental objects allow passage.

The game itself was subjective to a blind playtest at Pax, without the gamers being told it was by the Braid creator. "I don't really like feedback," he causes the audience to laugh, "[the pax playtest] was more of a reality check."

There is an obvious disconnect between the puzzles and the setting. Jonathan insists that this is in keeping with the fiction of the story as well as providing a fair interface for players. "In a game like Myst each machine is disguised to go with the scenery so the player spends ages just trying to figure out how the buttons move or if its a screen or whatever."

Following this I had the honour of being invited to a presentation by some students and ex-students from Newport Uni's games design course. Having now witnessed Jonathan Blow describe twice the process of educating a player about the rules of a game it's encouraging to see that these less-experienced graduates have taken on board the same skills. The beta level for Qube (available here at the time of writing) demonstrates this same, structured, non-narrative tutorial guiding the player through the individual rules and behaviours before attempting to combine them and looks very polished indeed at this stage.

There were a range of games displayed by the students. Hannah and Harry's ColourRunners (which I talked about on Wednesday) were there again and another group of students who are turning their final year project into a commercial product. It's encouraging to hear how much support Microsoft have provided to these under-graduates, providing access to hardware and dev kits for the at-the-time unreleased Windows 7 Mobiles.

Speaking of new Microsoft hardware...

GameCity5 - Day One - The Braid Director's Commentary

I have to start by being completely up-front about the fact that I haven't played Braid. I had assumed it had only been released on XBox 360 and PC but in searching to read about it in advance of GameCity, I find that it is out for PSN. Oh dear.

Nevertheless this session with the creator of the game, Jonathon Blow, in which he presents the game live via a project and talks the audience through elements of creating the game was absolutely fascinating and compelling from start to finish.

What is eminently clear from the presentation and the Q&A is that Jonathon is an enormously smart individual. From the way he tells it, the technology of the game was comparatively simple to put together but working out the laws of the universe in the game, how time travels through it and how this impacts upon the puzzles was of prime importance. At times, the discussion bordered on the philosophical. "There's been more focus on the text than the game design," he explains, "so I'm going to take about the design."

The assembled crowd is clearly in awe of the achievement represented by this game and the audience hangs on every word. It's clear that Jonathan is among like-minded people; there's a lovely moment where the Braid process hangs and he sheepishly calls up Task Manager to kill the process, muttering sarcastically when Windows asks if he is sure. This is followed by a knowing ripple of laughter from the audience and a recognition that even a feted indie developer is beset by the same problems as the rest of us.

I'll present some of my highlights of the presentation, because it was really great to be there and I really felt this session deserved its own blog entry:

A "Mario clone with crappy graphics"?

Jonathan started by presenting his original prototype - produced in 9 days. He described it as looking like "a Mario clone with crappy graphics" and highlighted that one of its failings was that the USP of the time-manipulation wasn't introduced until world 2, meaning that only the most enthusiastic player would make it past the initial levels to the really interesting features. This was why the final Braid would introduce the time-control from early on.

It interested me that he highlighted how quickly the player is in control. The opening title screen has the player able to move the character around and take part. Jonathan felt it was essential that the player would be straight in. This echoed, to me, what Jonathan Smith (Travellers' Tales) has said earlier about Lego Star Wars, that one of the key features was getting the players in and messing about as soon as possible - are people put off games by the layers of introduction, menus and tutorials that front so many games as standard?

He also showed how puzzles were framed, in the sense that he wanted to try and make sure all puzzle elements were on-screen at the same time and that where they were not, this was a deliberate choice. In fact, he said that with three years polishing, "everything" in the game that was present or not present was for a definite reason. (although he debunked any theories about the cloud in the final screen representing some kind of secret, saying "I just wanted a vantage point to look down on my castle".

The presenter is pleasantly candid about mistakes, pointing out sections of the game where a particular puzzle design or enemy placement did not sit well with him and that he would remove if given the chance to do the game over. Although, he defends some of the more controversial levels that he describes as "not fun" for a lot of players, saying that the "not fun" levels were kept that way for a purpose and to provide the game with depth and make it a more complete experience.

And how does a game like this get created? There's a clue when Jonathan is asked "at what point did you know you were committing to the game for three years" and he answers "I didn't ... honestly two years in I was convinced i was always 6 weeks away from completion." As each new element got refined and improved it provided inspiration for new ideas and possibilities and the game kept growing. Many game elements came from studying the code itself, for example time-insensitive objects.

In conclusion...

The opening day of GameCity 5 was, for me, packed with content, interesting discussions and chances to meet people as enthusiastic about games as me.

Now it's late. My writing is probably deteriorating by the word and I shall be doing it all again tomorrow. See you then!

Photo used courtesy of GameCity with thanks