As I’m sure you all have figured out by now, progress on the visual novel has been slow. Fred has mentioned before how much bigger the project seems now that he’s working on it than when he first started laying out the idea to get ready for the kickstarter.
He has done a lot of work. The early comics required a lot of writing to adapt them to the visual novel format. He has created the thumbnail storyboards for the Piro canon route in book 1 that I showed in a post last month. He’s plotted the Piro route and story branches out through the end of book 3. And he’s done artwork and put together the full game package for the early portion he showed in that demo video.
In spite of all that work, it’s still going too slowly. In these kinds of situations, Fred’s reaction tends to be to put his head down, shut up, and work harder. That’s admirable in a lot of situations. But if you need to get your boat across the atlantic ocean, putting your head down and paddling harder probably still isn’t going to get the job done.
Part of the problem is that the Megatokyo webcomic is really a one-man operation. Fred does it all, especially since he bought out Rodney. He’s not used to working with a team, and he’s used to having the total control that comes with doing it yourself. So while he has a team for the visual novel, he’s still been trying to do too much of it himself — the planning, the writing, the artwork, and all the coding in ren’py. It’s the approach he’s comfortable with, but it’s not working fast enough for the visual novel.
A couple of weeks ago he decided to make a change, and brought me and TheShad0w [John Hoffman] onto the team to help. I am now the ren’py script monkey. I am taking the original comic art and dialogue, using Fred’s thumbnail storyboards for direction, and trying to code that into ren’py in a way that fits the visual novel format. I’m trying to do it as close as I can to how I think Fred would want it. This is all work Fred was trying to do himself.
Shad0w has been working on tooling and some of the trickier ren’py bits. He wrote a script to pull and format all the canon dialogue from the Megatokyo website (where it powers the search ninjas) to save rekeying. He wrote scripts to generate placeholder character images so we can see their position on the screen, and torso, head, eye, and mouth positions without waiting for Fred to finish the actual artwork. That makes a huge difference in debugging the scripts and deciding what artwork Fred needs to draw. And he has written a number of tricky ren’py bits like animating the coffee pot that Kimiko throws at Piro, and making it look like Seraphim is sitting on the visual novel dialogue window.
The idea is that we will take Fred’s initial direction, do most of the grunt work, and give him back a functional game. He will still need to do the real artwork and make script changes so it’s the way he wants it. It will still need his magic touch. But at least he’ll be starting from something he can see and play rather than just try to picture in his mind. If we do this right, he should have to do a whole lot less work.
That should help us deliver the visual novel faster, because Fred’s time is the limiting factor. Anything we can do to let him focus on the stuff that only he can do, and not spend time on stuff that someone else can do, is going to be a net gain.
In the last couple of weeks, Shad0w and I have finished the Piro canon route for book 1. Even with the temporary artwork, it’s a fun game so far. I think you are really going to like it. Fred is working on the Piro canon route thumbnails for book 2 this weekend, and we will start on that next. Then Fred will start following behind with the script polishing and artwork.
I don’t yet have a feel for how much faster the game will come with these changes. But I think we’re all feeling a little more optimistic. We may not yet have fired up the outboard motor on our boat, but we’ve at least raised a sail. With a little luck and the right wind, hopefully we’ll reach shore with a finished game that everyone will enjoy.
“Daunting” is an excellent word to describe creating a VN. Having the process last year, with just myself and one colleague, it is all to easy to fall into the slough of despond. Setting tight and arbitrary completion dates helped, but all too often I’d be staring out the window wondering, “what happens next?” or hitting refresh over and over while I waited for someone on the LemmaSoft forum to answer my “how do I code this?!” question (helpful and polite bunch, by the way).
This move towards getting the basic structure down, with image and text place-holders as necessary, sounds good: it creates a tangible (so to speak) product, and gives everyone something to work on and from. I hope y’all continue to make progress.
Oops. Sorry. Wish I could edit that.