The Act 3 Countdown

March 15, 2019

Back at the beginning of February I wrote about completing the critical path implementation of Act 2 of my upcoming Sleuthhounds: Cruise game. At that time I made a rough guestimate that Act 3, which is somewhat smaller than Act 2, would be to the same state of development by about mid-March. At that point I couldn’t do much more than make an educated guess as I hadn’t fully worked out the major plot points for Act 3, much less the details of the puzzle design. Since then, I’ve been working on Act 3, doing a bit of implementation and a bit of planning. I know now the breakdown for the major scenarios in the act and have enough sorted out to take a stab at more refined estimates.

Task Description Estimate (days)
Cabin Investigations Searching for important clues in cabins. 4.5
Disguise Assembling and using a disguise in a particular puzzle change. 3.0
Interviews Interviewing assorted suspects. 3.75
Miscellaneous Various scenarios and housekeeping. 9.75
Suspect Investigation Interacting with various suspects as part of the investigation. 3.75
Timeline Ampson timeline construction puzzle. 1.5
Total 26.25

The biggest chunk of time in the above is the Miscellaneous category. In truth, I have this category, as well as all the others, broken down into finer detail for my own planning purposes. However, I’ve rolled the individual line items together into these higher level categories to avoid anything too spoiler-ific for the game in this blog post.

Software development is notoriously difficult to estimate and this applies for computer games as well. Over the years I’ve come to understand that by applying a certain scaling factor to my initial estimates, I’m usually quite accurate for what the actual development entails. Consequently, each of the numbers in the table already incorporates the scaling factor. Between the hours I put in on weekdays and the hours I put in on weekends I’m expecting this development to take until roughly the end of the second week of April.

In developing these estimates, I drew the numbers together without considering my projected end date from that earlier Act 2 blog post. In fact, I had to go back to that blog post itself to see what I had said regarding a rough completion date. Understandably, I was quite pleased that the more detailed estimates here coincided closely with my very rough, high level guess from before.

The useful thing about having a development timetable, such as this one, is that it makes it much more apparent if development is going to schedule or not. It’s been said that work expands to fill the time available and if there is no calendar to work towards then work can just keep on expanding and expanding. Something of this sort happened during the development of Act 2, in fact. With this timetable in place the rest of the development for Act 3, which has been quite steady so far, should continue to be focused. Whether it comes in as per the exact estimates remains to be seen.