A Lull in Development…
So I’ve picked the game concept that I’m going to fully develop. It’s a very close approximation of one of the game concepts I posted publicly last week, though I’m sure it’ll evolve somewhat as it develops.
Having firmly made this decision, and also decided on an art theme for the game, I’m now twiddling my thumbs a bit waiting for my artist to finish the first robust set of art. There are, I suppose, things I could be doing, game-wise, at the moment. But really, once you’ve found a game mechanic for a casual game, it’s mostly about art implementation (with corresponding special effects), UI implementation, and gameplay tweaking. The latter is kind of hard to do until you get the main game working with reasonably polished art.
So I’m probably sidelined for a week or so. Which I suppose is ok – I’m taking a 4-day weekend this weekend, and I’ll fill in some of the rest of the time by doing my taxes and other odds and ends. Still, I always feel a bit odd when I’ve got nothing productive to work on, biz-wise.
I think this game will go very fast – I’m hoping to be out the door in 3 months or less from now (after spending the last 1.5-2 months concepting and goofing off a bit). Bonnie’s Bookstore took 6.5-7 months to go out the door, with a shorter concepting stage – about 2 weeks. But I was also doing my first casual game, working with and enhancing a very rough framework.
Bonnie’s Bookstore Codebase (.c files only, excluding utilities)
Framework: 345KB
Game: 400KB
I’m reusing all the framework, and will probably only add another ~50KB of new code to it. The game itself will obviously have a lot of new code, but also some re-used code, and I’ve already got the prototype working (though a lot of that code is slapdash and needs a re-write). But anyways, I’m already at 167KB for the game prototype, and will probably only hit about 400KB again, meaning I have perhaps 250-300KB of code left to write.
When I was in full-on crunch mode at PopTop, I could produce 100-120KB of code per week, but 60KB was a more normal output (and that’s counting spending some time tending to managerial/design tasks). I doubt I’ll be even that productive on my next game, as I work fewer hours and perhaps spend a greater portion of my time doing design and graphics tinkering. Maybe I’ll do 45KB of code/week, which means that this new game could be done (at least to late-beta stage), in about 6 weeks of work, from the point that the art is ready to work with.
We shall see…
March 1st, 2006 at 4:07 am
I was thinking 100kb per week is ridiculously huge amount of code per week, and that you must be bullshitting, and then I realized that you are not talking in LOC.
OTOH 100,000 chars divided by 7 is 14kchars per day. Assuming 100 chars per line, that’s 140 lines/day, which is more reasonable.
March 1st, 2006 at 9:09 am
I usually assume ~30 bytes/LOC.
That’s a crude assumption, based on looking at various large .C files I’ve written over the years. The LOC is simply the number of lines in the file (measured by any standard text editor), NOT the number of function points or anything like that.
It varies a bit for me – probably the more accurate number is somewhere in the 31-34 bytes/LOC range, but 30 makes the math easy.
So when I have one of those 100K weeks (again, rare, usually during 7 day/week crunch mode), I’m doing about 3,300 lines/week, or about 470 lines/day. If you’re writing relatively simple code, it is possible to bang out that much.
March 3rd, 2006 at 5:05 pm
mmmm… bbq is good!
did i guess right?
m3mnoch.
March 4th, 2006 at 2:45 pm
That’s just new code right? I sometimes used a mixture of new and old code where I had to manage file access and routine stuff.
Though my typing speed is inversely proportional to its accuracy so going fast for me just means more compiles.
March 6th, 2006 at 6:21 pm
When I talk about ‘new’ code in the above, generally it means fully new – not cut, pasted and modified, though I do that sometimes.
March 7th, 2006 at 12:42 pm
Hey Phil, I really liked this post. This timeline is great for a second game when you have a lot of the framework in place. If this had been your first title and you had no framework, how long would you typically expect a project to take?
March 7th, 2006 at 1:09 pm
Nick – it really depends on how you’d go about getting a framework, and your experience level. Using a 3rd party framework (i.e. PopCap’s) would probably be faster than rolling your own, but that would depend on how closely the framework in question fit your needs. Off-hand, for me personally, ditching my current framework and using PopCap’s framework would add 2-3 weeks. Ditching my current framework and rolling a new one from scratch might add 4-6 weeks.