Nov

Dev Diary 13: Milestone 0 Review

Yac on November 4th, 2008

November 1^st was the deadline for my first milestone (aptly named Milestone Zero), and to “commemorate” the event, I did a review of the project’s progress. I originally wrote this assessment for myself, so it’s a little dry. But I might as well just share it with everyone so you can see what’s gone right and wrong so far.

3-month Project Review – Milestone 0

The major goals set out for this milestone were:

Develop and prove the core technology.
Settle on the design.
Submit a demo entry to the IGF.

I have accomplished only the first goal, and failed on the second one leading to the obviously failed third goal – if there is no game, there is nothing to submit to the IGF.

Along with those goals, in late August I laid out project estimates of the major tasks I needed to accomplish to reach the goal of submitting an IGF entry. Those sets of tasks were further assigned A, B, or C priority to denote their importance. The A’s being absolutely crucial, the B’s less so, and the C’s being pure polish. I laid out this plan, reviewed my estimates and then did not change it again for the next 10 weeks. I wanted to see how I performed against longer term estimates.

I accomplished 80% of my A tasks:

Implemented the core physics technology, which is now working and reliable.
Implemented a level editor.
Implemented a simple GUI.

I didn’t complete my last major A task, which was finishing the core design of the game.

I only accomplished one B task, optimizing the core physics tech (which now runs fast enough for my uses). The other (unfinished) B tasks, were related to the visual quality of the rendering (supporting textured objects, shaders, etc).

I didn’t even touch the C tasks, which comprised audio support, and packaging for release, video capture, and publicity asset creation (screenshots, trailers, etc). These tasks would only have made sense for an IGF submission.

Conclusions on Task Performance

I am satisfied with my progress against this list, and somewhat surprised at the amount I accomplished. When I created the tasks, I assigned an optimistic and a pessimistic estimate for each task. I also over-tasked myself, not because I thought I could get it all done, but I wanted to list out everything I wanted done for an IGF demo. I had 50 man-days of resources, and I had 56 (optimistic) and 88 (pessimistic) days of work scheduled.

The major disappointment is not accomplishing the design task, although not for a lack of trying. The lack of major improvement in visual quality is also a negative, although not as important as the design.

Overall, I tracked at around 75% in terms of efficiency, a number I am very happy with given my past experiences with sprints, project estimates and planning.

Areas that Suffered (and Why?)

Design:

Losing a designer pushed things back a bit.
The tech tasks took longer than expected.
Relative unsureness on my part in the design field has held me back.

Visual Quality:

Lower priority than the other major tasks (core tech and design). I don’t need the demo to look great until I have a fun prototype. Gameplay before graphics applies here and it would have been misuse of my resources to focus on this area at the expense of others.

Areas that were Successful (and Why?)

The core tech:

Pressure: Without this tech, there is no point in trying to create the game I want to make. If I couldn’t nail the tech, there was no point in continuing. I would have just killed this project and moved on to something new.
Most of my energy and worries was focused on this area.
I spent more time on it than planned. I worked on it till it was done.

Breakdown of Areas of Focus from the last 10 Week

Week 1: Core tech, camera.

Week 2: Core tech, architecture.

Week 3: Mostly holidays, some core tech.

Week 4: Math, architecture, some core tech.

Week 5: Box2d, core tech.

Week 6: Box2d, core tech.

Week 7: Core tech, editor, gui.

Week 8: Serialization, core tech, editor, admin.

Week 9: Core tech, editor.

Week 10: Editor, gameplay.

How Can I Improve?

I’m doing two major things to improve how I work. I am now going to work on monthly milestones (actually, 4-week milestones for consistency’s sake), and review the overall project status at each stage. There is a still a long challenging road ahead, but I also don’t want to be afraid to kill or shelve this project if it is the most sensible move. I especially don’t want to find myself 6 months further down the line with no clear end in sight.

The second thing is to continuously address the highest risk areas on the project. This is something I’ve done in an ad-hoc way until, but that I want to formalize. For every milestone, I will prioritize tasks so that I address my highest risk areas first. The design is the current biggest unknown (now that tech is no longer an issue). Design is therefore my highest priority.

Work Done Last Week

I finished up some tasks on the editor, and started focusing heavily on the design, experimenting with a few different ideas (nothing’s stuck yet). I ended up doing the project review over the week-end.

Work Planned This Week

I’m going to focus mostly on experimenting with various design ideas this week and I’ll try not to get stressed over the whole process. Thank you to my good friend Andy for reminding that designing the game is supposed to be funnest part of development. I sometimes forget to enjoy the process.

I may dabble a little bit in the graphics side of things. They say that once you have a crate in, it’s officially a game. I’ll need texture support for that.