Latest estimate: this Saturday (December 27th.)

Sorry for the delays. I'm genuinely sorry. I didn't post yesterday because (a) I was working so much on the game, and (b) I'm frankly embarrassed. I feel really bad about not delivering on promises and want to hide and not show my face until the game is actually, genuinely, really really uploaded and ready.

I promised the game would be ready by December 15th, and it wasn't. And now every day all I can write in the blog is "nearly ready" and "nearly ready, honest" again. Hopefully most people are too busy with Christmas to won't notice.

The game was pretty much ready to upload last Monday (the 15th) but "pretty much" just means "no obvious bugs." And "no obvious bugs" means "lots of little bugs if you look closely or do something unexpected." By "bugs" I mean irritating little things, like clues that aren't clear, typos, characters who turn in a circle before leaving a room, etc. So I've been fixing those things, and every code change risks creating new bugs - for example, I made it so you can choose "open" to open a new game even when the intro is playing. But that required complicated code behind the scenes (since you're not supposed to be able to interact with the game during a cutscene), and that stopped a later scene from working properly. That's fixed now. But there are hundreds of little things like that, some take ten seconds to fix, some take ten hours. Some days I have twn hours to work on the game, some days only two.

Remember I said the game was 99.9% finished? There's a saying in programming

"The first ninety percent of the task takes ninety percent of the time, and the last ten percent takes the other ninety percent."

This is a variation on what is now called Hofstadter's Law:

It always takes longer than you expect, even when you take Hofstadter's Law into account.

Hofstadter is right. When planning this game I allowed several months for the unexpected. But the unexpected is always so darned unpredictable. But it's good to know I'm not uniquely incompetent - these things happen to everyone.

I was relying on last weekend to make the final changes because I had three days off work. And those were the three days that we had major power cuts. I counted about fifteen power cuts, ranging from two seconds to twelve hours. So in the darkness I wrapped presents, then found a torch and started to design Dante's Inferno on paper. Dante's Inferno is coming along well. :)

By yesterday I'd made all the latest changes from the final test, and was about to upload the game when I thought "one last run through just in case." A good job I did, as some of the latest changes caused new problems. Nothing major, but I really need to play the whole game again from start to finish. That's what I'm doing now, making minor fixes along the way.

This is really the polishing stage. Whenever I find a clue that seems too hard I change it, and if a character still doesn't look right I change that as well. This is time well spent, and the game will be better because of it. My favorite change is the "poke with a stick" section. My youngest daughter suggested it. In the sewers, when you get a stick, she said you should be able to poke things with it. And so now you can, with various accompanying sound effects and comments. She also said you need to see the bricks falling when they fall, so I added that too. (She also wanted bats to fly in the chimney, but that would have taken too long to animate.) These are all just little things, but they should all add up to make the game a more satisfying experience. Most of my time is spent on fixing background irritations, like the character-turns-in-a-circle that I mentioned before (that took six hours to fix), and the not-always-reliable way that characters entered rooms, so I re-wrote AGS's built in "place on nearest floor" code and it seems to work better now.

As I've said many times before, the same code will be used in all the games, so future games won't take this long to make. Indeed, the whole point of having the code "just right" is so future games are much faster to produce. If nothing else, I'll know what's easy to design and what to avoid at an early stage. No doubt Dante will raise a few new challenges (regarding jumping between games) but by the second half of 2009 (Apeiron) all unexpected delays would be ancient history.

Anyway, enough rambling. I'll have the final, final, final play through finished tomorrow or boxing day (Friday). Then I'll upload the game and check that it downloads properly from the commercial site that's hosting it. Then I'll make the free demo (which on past experience will take longer than expected, but IN THEORY should only take four or five hours to put together). Then upload the new design web site (which is already done offline), send out the "come and get it" email, send free copies to everyone who helped, and finally go to bed and never ever look back on 2008. 2009 will be so much easier.