For the past three years, Premier Press (now Course PTR) has held a party for their authors. This year was bigger than ever, as the shots below illustrate.

Emi Smith (Course) and Wolfgang Engel	Kevin
Bob Bates talking to Mitzi Koontz (Course)	Stacey Hiquet (Course), Kevin, Heather Hurley (Course), Dave, Emi, Jana Halford, Neal Halford
	Nick Robalik and Pam Crowley (Caligari)
Kristin Eisenzopf (Course)	Dave and Emi
We love the Gordon Biersch
Kevin at the buffet	Kristin and Kevin
Kelly Dempski and Mitzi	Stacy and Kevin
Kelly and Mitzi again

David Michael already hit on alot of this further down the page, but I figured I'd give my take on this session of the Serious Games Summit as well. The session's panel of moderator Noah Falstein (The Inspiracy), Randy Hinrichs (Microsoft Research), Doug Whatley (Breakaway Games), Andrew Kimball (QBInternational), and Chris Crawford (Erasamatron Project) attempted to answer the question, "What, if any, are the principles of game design that are critical to education and training?"

Doug Whatley kicked things off by mentioning how when developing "serious games" it's important to research the perspective of your game first before you start on the game's design. By perspective, he means narrowing the scope of the game world to serve a specific purpose in education and training. He also said that being interested in what you're teaching through the game helps in maintaining interest and quality throughout the project.

Chris Crawford then stepped in by saying that the simulation is not about material, as in a game, but rather that a simulation is about the student. An overemphasis on technical correctness is not required, nor is it good. All simulations are of a subjective nature from the user's perspective, and no good simulation is objective, meaning it's not the same experience for every user. The important part about simulations is getting the simulation's message (or training) across at a level the student understands. Crawford finished off with: "Technical accuracy can be found in a 'big heavy book somewhere'."

Randy Hinrichs followed up Crawford with a focus on instructional design. He said that we learn the fastest when all five senses are used in the learning process, so we should find ways to get as many senses involved in the playing experience as possible. He also mentioned that the user should build skills up throughout the game while giving reference points throughout the game so the user knows what to do. Enough feedback must be given with rewards at a proper pace. Most notably, the science of instruction should be applied.

Andrew Kimball decided to get more scientific in his views by bringing up the three types of learning that simulations help the end-user achieve: knowledge transfer (KT), skills transfer (ST), and attitude and beliefs transfer (ABT). KT involves getting the brain to struggle and forcing it to process information in whatever form may be appropriate. The more difficult the struggle the better. ST is about practice and feedback. Most computer simulations are not the best at ST learning from the standpoint that simulations are about practicing for the "real thing", and you can only achieve a certain level of practice in a simulation. A good example for ST is playing the violin. ABT occurs when a user has a jolt, or an "AHA!" moment. Computer simulations are terrific for ABT learning, but they must have a debriefing session after the learning experience. Debriefing was stressed as being very important, to the point that a simulation is only as good as its debrief.

Throughout this discussion, the topic of entertainment was brought up and whether or not a "serious game" can be entertaining. Everyone agreed that an entertaining serious game allows the user to focus more on the engagement, rather than on actually learning. A few research statistics were even brought into the discussion (no sources): 27% lecture retention after one day, as opposed to 82% interactive retention after one day. 4% lecture retention six weeks after the learning session, as opposed to 45% interactive retention after the learning session. Judging from these statistics, it's clear that interactive entertainment only increases the learning retention of end-users. Kimball also brought up that the purpose of learning in games is to get the maximum KT in the least amount of time.

At this point Chris Crawford brought up a few good points differentiating educational and training games from entertaining games. He started with a series of analogies:

Data vs Process

Goals vs Services

Assets vs Operations

Nouns vs Verbs

Facts vs Ideas

Knowledge vs Understanding

Everything on the left side of these analogies deals with data, while everything on the right side of these analogies deals with process. Crawford argued that simulations are about process, and that we should focus on the process instead of the data. He mentioned that the worst mistake we can make is to design a simulation and think about images, sounds, and other data-related components.

Client needs were also briefly mentioned. The consensus was that we need to caution against thinking what's the best game and what's the worst game when designing serious games. Rather, we need to think about implementing a game that serves its training purpose.

Many other good points were made throughout the panel:

Remember that simulations are about working in a real-world context

Multi-user games tend to be more effective for learning due to complex interactions with other users

New devices could help enhance learning with games

Don't be afraid to allow game design to evolve into other technologies

A user lacking understanding of the game's mechanics can help teach - i.e. make a virtual world, let the user play and learn the game's assumptions without any training or education on how to play the game

These next points were brought up when Falstein brought up the topic of serious games being "fun" and the importance of "fun" in serious games.

3 elements influence retention from e-learning:

fun - creates iteration of the game. It keeps the user coming back for more

relevancy - is the game relevant to the job?

internal performance reinforcement - does the user get feedback on mistakes and successes?

Without fun people might play once or not complete the game at all, thereby lowering retention

Possible the best set of points over the topic of fun in serious games were given by Chris Crawford. He said that the question of whether serious can be fun is metaphorical to "school is no fun", and "fun can't be educational". He gave an excellent example: Take a bunch of kittens playing in a room. The kittens play, have fun, and are learning the skill of the hunt. Going along this line, Crawford then posed the question, "Do we know that school is educational?" The audience responded with a slew of laughs, chuckles, and "OOOHHH"'s. Crawford ended his point by saying that we should give people the option to validate the proper rules that the game was intended to teach, rather than forcing those intended rules from the beginning of the person's playing experience.

I arrived here today after a grueling 8 hr drive, after 24 hrs of no sleep and a car crash (thankfully nobody was hurt!) I managed to drag myself up to the GDC to meet industry guru Dave Perry on his personal game industry experiences and snag some advice for those of us who are trying to break into the industry. Overall Dave gave me some excellent information and I'm looking forward to posting the interveiw later this week.

This year's Microsoft DirectX party was a bitchin' good time, I must admit. It was at Zoe's rather than the Agenda like years past, which was just across the street convieniently enough. The club was only one floor, but after I got in and explored I realized it pretty much equaled the space of Agenda's three floors. In fact, the club had three distinct sections with a bar in each. When you walked in it was like the dance club, with an open area for the dancing and tables scattered around the bar. The next room back had two pool tables, tables, and two demo kiosks from NVIDIA. One showed a cool shader demo, and the other had Halo. There was also a small patio for you to catch some fresh air... if there won't half a dozen people out there smoking at any given time. Oh well. Anyways the last room was nice and relatively quiet, great for actually having conversations, and it also had two kiosks, one with a demo of FarCry (which was looking awesome) and another with a demo of... uhmm.... *attempts to reference pic* rats I can't remember. I think I was focusing too much on FarCry to even care :P Anyways all these rooms were like the same size, so the club was indeed pretty big. Last year at the Agenda it was shoulder-to-shoulder packed, this year you could actually move around. I wish MS had thought to use Zoe's before this, and since we'll be in San Fran next year I guess it's a moot point. But still, my hat's off to MS for throwing a kick ass party.

Ack, how could I forget about the cool handouts? Well this year they had blue blinking LED rings that you could affix to just about anything. Being the swag-king, I of course managed to swipe two instead of just one. Though some guy had three... punk. Some hand them on their hands, others their ears, shirts, nose, hats... They were also giving out those green glowing NVIDIA sticky thingies like last year and lots and lots and lots of glo-necklaces. One guy managed to cover himself completely in glo-necklaces. I don't know why I didn't take a picture of him. D'oh.

Outside the nightclub waiting to get in	Here's the line 15 minutes before the party
These ladies were dressed for the occasion! (there were 4 of them)	All of their outfits not only glowed a-la Tron, but some of the lights were blinking too.
This was the dance area in the first room. The Glo-girls would take turns with swinging lightsticks up on the podium stage. I still don't know exactly what those guys with the laptops were doing - the audio booth was behind me	I forgot to turn the flash back on, but I thought this pic looked kinda cool anyways
Ah there we go. Nice smoke (courtesy of a smoke machine, not smokers) added to the ambiance. Hey lets play a game! Who can find Dustin Clingman?	Here is the second room, pool tables to my left, and you can just see the bar on the right
The shader demo kiosk, the Halo one was to my right	The third room with the two other kiosks in the front. The bar is to my left
Another Glo-girl doing her thing	Ditto. This is all they do. They're a group that does parties like this - all the glo-necklaces and bracelets and stuff were from them
Game developes can dance!!

Yeah part 2 is above part 1. You've gotta read this stuff bottom-to-top if you wanna see it in chronological order.

After a change of laptop and several more minutes with the A/V guys, I walked out before Scott Kim got his speech out. According to Ron, they did eventually get it working but the presentation was underwhelming and the PowerPoint only half-worked because the QuickTime videos wouldn't play.

There were a couple of standouts in the later presentations. One was an excellent pair of presentations on "A Casual Game that Succeeded" (here) and "A Casual Game that Failed" (here). The producers of both products were the presenters, so you got a good idea of what went right and what went wrong with both projects. Also good was a presentation about the business of selling casual games, mostly for cellphones. The cellphone game market is a huge mess right now, and it's largely because there are three forces that can't agree on a business model

• Cellphone makers: Move as many shiny new plastic phones as possible
• Cellphone service providers: Sell as much airtime as possible, leaning towards what makes the most money (ringtones and instant messages)
• Game developers: Sell as many games as possible

As you can see, this isn't exactly a recipe for success. The summit ended with a discussion of technologies for casual games, centering on the expected targets: Flash, Director, Java, C++, and to a lesser extent WildTangent and .NET.

Casual Games Summit

John already posted a summary of this session, but I figured my notes might provide a supplement to his. There is some repeated information. Feel free to sue.

Steve Meretzky presented an overview of casual games, beginning with several definitions:

• A game for which the intended audience is someone for whom gaming is not a primary area of interest.
• A game where you can learn the rules and be up and playing in a minute or less.

Characteristics of casual games:

• Low barrier to entry.
• Forgiving/friendly and helpful to the player.
• Short playing time.
• Highly replayable.
• Convenient/quick-starting.
• Addicting.
• Inexpensive.

Having defined casual games, Mr. Meretzky then proceeded to outline why developers should be interested in such games:

• Large market/audience.
• Money to be made (the audience will pay for what it likes).
• Entry costs are still manageable.

A short history lesson showed that, in the beginning, most games were what would be considered casual games today: Pong, Pac-man, etc. What are considered "core games", though, began dominating the game market in the 1980's.

Causal game audience:

• Most players are female, and their favorite types of games tend to be word games and puzzle games.
• More likely to be older. Players over the age of 50 are the fastest growing segment of game players (by age).
• More international. Word games, though, tend to be language-specific.

Kent Quirk's presentation after lunch, describing the failure of Cognitoy's attempt at the "perfect casual game", was entertaining and enlightening. He talked about how they inadvertently chose to combine 2 game concepts that didn't mesh as well as expected, comparing the result to "Terminator meets Fried Green Tomatoes".

Indie Perspective, by DavidRM: Obviously, the Big Boys are taking notice of the casual game market--and not just in the "mobile space". While this "proves" the market, it also demonstrates that indies need to get their act together and realize that the days of easy access to this market are fast fading. Several times it was mentioned that it had proven valuable to use a free, Web-based game, often hosted by a portal, to cross-sell a downloadable version of the game. This may be a good way for indies to get the best of both worlds: licensing to a portal, with its large audience, and generating sales directly to players.

Serious Games Summit Panel
Rules for Serious Games

The panel discussion was interesting, and even entertaining. Here are my notes, organized by panelist.

Noah Falstein (moderator)

Noah talked about his experience playing the game SCRAM, by Chris Crawford, which simulates a nuclear reactor. The game was fun, but to excel at the game required that he learn about nuclear reactors. "You make a fun game," he said, "but the informatin you [the player] need is real world information."

Randy Hinrichs

The content of serious games must be engaging, and must have a context, with the player involved in a relevant environment.

Doug Whatley

You can't design a game and then find a way to teach. You have to know what you're trying to teach and design to that.

Andrew Kimball

The type of game depends on your learning objectives. There are 3 types of learning objective:

• knowledge transfer - happens through memorization and repetition.
• skill transfer - practice and feedback are vital; this is hard to do online/with a computer.
• shifting of attitudes and beliefs - requires a mental or spiritual "jolt".

Referencing the shifting of beliefs, Mr. Kimball said, "A simulation [providing a jolt] is only as good as the debriefing" that follows it. That is to say, what was learned needed to be put into context, and what was learn shown to the player.

Mr. Kimball also talked about discovering "game frames" that work and can be used over and over. Storytelling, for example, is one of the most effective methods of self-examination ("with proper debriefing").

When asked how important fun was to learning, Mr. Kimball stated that without fun, retetion from e-learning goes "way down".

"Anything that draws you out of your head is entertainment." Learning, however, requires that you be "in your head".

Chris Crawford

"Technical correctness," Mr. Crawford said, "is less important than presenting a coherent whole." In other words, the learning experience offered by the game should not be so focused on being absolutely correct that it trumps the learning process. Learning requires a presentation that is consistent with itself.

Simulation is about process, not about data, operations not assets, cycles not bits, and so on. "Simulation works when you focus on the process."

When discussing how important fun was to learning, Mr. Crawford pointed out that play, with its associated fun, is the original education technology. "Do we know that school is educational?" he asked.

I find it hard to believe we've forgotten to post about this. Last year there was rumor of moving the GDC 04 to Las Vegas. While Alan Yu shot that down when I talked to him last year (the interview never made it to press), there was still some lingering doubt I felt as to what would happen to the GDC's location in the future. This year I guess they've decided we've reached critical mass for the San Jose Convention Center because next year we'll all be packing up and heading to.... San Francisco! It's true! The conference guides do not lie. How about them apples??

In the morning session of the Advanced OpenGL Programming tutorial, in addition to going over the details of the OpenGL Shading Language, they covered the current plan for OpenGL 2.0.

At this point, it seems likely (though obviously not certain) that the OpenGL 2.0 spec will be approved by SIGGRAPH in August. Although the features to be included in 2.0 are still being discussed, and changes may very well be made, the following features will most likely be in 2.0:

• OpenGL Shading Language
• Floating point pipeline - full floating point support throughout the pipeline, and values that were previous clamped [0,1] are no longer necessarily clamped as controlled by the program
• Pixel buffer objects - do for pixels what VBO's do for vertices, to be used with glReadPixels, glTexImage2D, and so on
• Sprites promoted to core OpenGL functionality
• Texture rectangles (however, this has not been finalized)
• Non-power of 2 textures (may be limited to 2D textures)
• Enhancements to accumulation buffer - modern hardware now actually support accumulation buffer
• Multiple draw buffers based on ATI_draw_buffers extension, similar to multiple render targets in Direct3D
• Two-sided stencil - allow shadow volumes in one pass
• Greater control over blending
• ARB_extension_query - adds a single function: glIsExtensionSupported. Allows you to check the extension without querying the string

The following features have been discussed, but are unlikely to be included at this point. To reiterate, until the spec is actually approved, all of this is subject to change.

• Uber/Superbuffers - not going to be ready in time
• Texture mirror clamp - not useful enough to be in the core
• ARB_vertex_program and ARB_fragment_program - since the focus is on the high-level shading language

I was originally planning on attending a tutorial on leveraging development deals to build better value in your studio but I got distracted by Microsoft giving away free stuff. Blast that Microsoft! Anyways even so I think I enjoyed this tutorial more than I would have the other one.

The first half of the tutorial was an overview of programming properly for WindowsXP and handling issues such as robustness, security, fast user switching, multi-users, etc. In a huge twist of irony, as Kev Gee was discussing WindowsXP reliability issues, the computer crashed. Well, to be fair it was more of a computer crash than an OS crash, because the entire computer just shut off. But it was still funny anyways.

The second half (which I didn't hang around for because the topics presented were beyond me) split the gathering into two audiences by partitioning the room and one side got a Production view of HLSL and associated technologies while the other side got a Developer view of HLSL and associated tech. Obviously the developer view was more in depth and technical and how-to whilst the production view was more along the lines of how shaders and things work in games.

One of the cooler things presented was the PIX profiler, currently in Beta 2. It's been used extensively for XBox, no doubt due to the much stricter performance and memory requirements of the console, and now PC developers are being encouraged to pick it up. It profiles D3D with myraids of options available to tease out exactly what's bogging down your game and causing those frame rates to suffer.

There's nothing much more I can say other than this brief overview due to the scope of the day's tutorial (much bigger and more meaty than yesterday or sure). However they did provide all the lecture slides on CD in the nifty backpack they gave out, so I've posted them online and you can download the zip file here

A partition will come out to seperate the meeting room during the second session, but for now the two screens are synched up	This is the Developer side... yea!!
Mmmm free backpack with lotsa pockets (more on the side, flap, and back)	You can carry this in three ways. 1: by the handle on top 2: by a shoulder strap tucked away in the main pocket 3: as a backpack with two shouldet straps tucked away in a rear pocket. Smooooth

Much more up my alley than the 3D browser-games presentation yesterday was today's "summit" on casual games, defined as games playable and played by people who normally don't play games. They presented several examples of card games, ranging from the successful but very sparse Pretty Good Solitaire to the successful and very beautiful Hardwood Solitaire. They also covered game shows, trivia games, and sports games in addition to the standard card games.

After 11:00 we had a presentation on mini-games embedded in larger games. It was interesting, but I generally despise mini-games embedded in larger games. From my experience releasing game packs, I can ensure you that you will never be able to make a set of several games that everyone will like. You will inevitably end up with a few games that many will love and a few that many will hate, and that list will be different for every person who plays them. Forcing a user who dislikes Tetris-knockoffs to get bonus points by slogging through a Tetris knockoff (or even worse, forcing him to beat the game before it will continue the "main" game) will virtually guarantee that your game will be shelved before it is finished. Making it difficult for a user to reach games he likes and forcing him to occasionally play games he dislikes is not the model for a successful game.

On to another rant. Scott Kim, of course, was a presenter. Scott Kim, of course, is unable to get his presentation materials to show up on the projector. Every time I've seen a Scott Kim presentation, there's been much hair-pulling and screaming while getting his stuff to show up on the screen. I don't wanna pick on the guy, but if you're presenting to a couple-hundred people who paid over $1000 to see you, you could work a bit harder on preparation.

Business stuff is after lunch. It'll be interesting to see what makes a casual success story and what makes a casual failure.

So okay okay. We got caught. Drat! It was only the second day too! Why couldn't we have gotten caught like on Friday or something? I am now cowered, and will be purchasing my ticket every day for the remainder of the week. I repent! I repent! (I figure there's a slim chance the tran-cop will ever read this but I can always try)

Anyways he yanked us off the train at the next stop, walked to the ticket machine and placed his finger on the $20 ticket symbol. Ack! Luckily he just meant to buy a ticket, not necessarily a ticket that added up to $20. So Ron and I got a $1.50 one-ride ticket. John will be collecting bets for our return trip tonight, so be sure to send him lots and lots of emails with your bets. Lots.

In other news, Just as slim odds caught us, slim odds also presented us with Nick this day, and the odds weren't in his favor either. He got caught at the Convention Center station... but he didn't have to buy a ticket. Punk.

From the Day's Inn, there are two ways to get to the convention center --the GDC bus or the San Jose Light Rail. The GDC bus is free but doesn't run on a very good schedule. The Light Rail runs all day but costs a couple of bucks to ride. The LR's ticketing basically runs on the honor system. It's a free ride as long as you don't get caught by the train-cop who's occasionally checking tickets.

And guess who got caught this morning?

Honest ticket-purchasers John and DavidRM continued on to the convention while Ron and Gaiiden stayed behind to get yelled at by the tran-cop. Hopefully they'll post their own adventures so we can find out if they got fined. Gamedev.net's ad-hoc Light Rail Delinquent Pool odds favor Drew getting the fine but Ron geting sympathy for crying like a woman.

Speaking of odds, the line is now 4-1 that Nick shows up today and 20-1 that he actually writes anything. Stay tuned!