GK3 Designer Diaries - by Jane Jensen
March 3, 1998
Well, remember that stuff about the sword of Damacles in the first entry? It materialized again sometime in December. At least it's getting easier to recognize it as a normal part of the production process. The problem: We're eight months from shipping and halfway through our development cycle. We have enough of a handle now to know how long things take and exactly how much there is to do. Unfortunately, we're not halfway through the amount of work to be done, so an adjustment is necessary. Of the three - schedule, budget, or project design - something has to give. Since this is a business, it's usually the project that must adjust.
With GK3 our critical path (area of development that's most tightly scheduled) is animation. 3D animation is a much different process than 2D animation. When a character interacts with a piece of furniture, an object, or (god forbid) other characters, there are a lot more man-hours involved in making that work than there was in the 2D realm. Each object is a full 3D "thing" with coordinates and arcs and so forth. Making them match up together and look natural in conjunction is very tricky.
Therefore, I've spent the past six weeks going through the scripts once again in order to rethink animations. It's a tedious process, but as a designer, you're stuck with this kind of task because you know the interrelations of the scenes and have the global view of the flow of the play that no one else on the team has. You know how each modification will affect the product. And also it becomes a choice; would you rather have someone else change the game or do it yourself?
Every project I've done has gone through this, and part of the reason is because technology never stands still. There's always a learning curve to figure out what the possibilities and limitations are in each new project. If we had done GK3 in video again, we would have been pros at the process and been able to produce the game faster and known all the pitfalls up front. But that's not the way the industry works. And because the design is due in well before the rest of the team ramps up, that means often having to "correct" some assumptions down the road once the real rules of the process become apparent.
To give you an example, having a character move a large trunk across the room or come downstairs with an ice bucket are simple things in the old SCI 2D world. But in the new G-engine, they involve a massive amount of animation time (each object has its own animation count as well as ego, plus extra time to coordinate the two). So much time, in fact, that it is a better trade-off to forget about the ice bucket or have the trunk already in the correct position (assuming these aren't critical parts of the story) and "spend" that animation time on a part of the game that's more integral to the plot or more fun for the gamer.
The good news is, I have survived the process and so has the product's integrity. Hopefully, it will be my final pass on the scripts. We may also be hiring an additional animator to help us produce more, so the company has been willing to meet us partway. That's a good thing.
In other news, we've been showing the game to more press recently and have started to get some preview coverage. We will be meeting with marketing and creative services next week to begin brainstorming about the box. Meanwhile the team crunches on, still working on the first few time blocks. There's not really enough of a game for me to spend much time play testing yet, but hopefully in another two months or so that will be my primary job.
The latest GK3 web site update reveals more about the theme and setting of the plot and about the casting, so instead of going into detail, I'll send you in that direction if you're curious: www.sierra.com/titles/gk3. Until next time....
Last update: October 24, 2007
Delicious 
Digg
Facebook
Reddit
StumbleUpon