State Of The Game #3
It's been a while since I wrote a state of the game post. It was supposed to be done back in Jan, but then things happened.
I'm finding it harder and harder to write blog posts because the most interesting things happening right now on the project would be profound spoilers, but we shall forge forward, none-the-less, and-all-that.
On to the state-of-the-game.
It's been a long road. We started working on this little game we call Thimbleweed Park a year and three months ago and I feel we've grown to be a nice little family, not just the team, but also everyone on the blog. The project progressed pretty peachy until October, then the holidays showed up and it felt like we hit a snag. We had a lot of drive coming out of our first sprint, but then we puttered along after that.
Not sure exactly the cause, maybe a little burn out, accompanied with feeling a little overwhelmed with the monster of a game we'd created, but persevere we will and did.
Just about the time were were getting over post-holiday focus back, GDC showed up. Our original plan was to have an open-to-the-public party where people would get to play the game, but as we started making plans, we realized we just weren't to a place where we'd be happy with the game we had to show. On top of that, everyone was very busy, and organizing something like that was just too much work. We gave it a good try.
So we opted to do a press only showing of the game. When you're demoing to the press, it's a lot easier to do hand waving and divert attention from areas of the game that aren't finished or even slightly broken. The press are used to seeing games in this state and generally know how to interpret and project what they are seeing. This is fine for a first look, but not for a preview or a review, but that's all we were doing.
Prepping for the press demo took a lot more work than we anticipated. It hijacked the sprint we should have been doing as we instead entered a phase of polishing. None of it was lost work, it all had to be done anyway, it was just done out of order and distracted our focus.
Polishing included adding special case animations that we often leave until the end (in case the puzzle changes and it's not needed anymore), touch-ups on art, making sure all the verbs that make sense don't respond with "that doesn't seem to work." We also added a lot of ambient background animation, like waving flags and twinkling stars, so the rooms had a static life to them.
For the demo, we decided to show an abbreviated opening to the game and then jump to the place where the Ransome flashback happens and began polishing and testing those areas.
Around this time, we also decided to switch our backer system to PledgeManager so backers could upgrade their pledges. What started out as a quick few-day project had spanned into weeks, all the while we were trying to get ready for GDC.
And... in the middle of all this, we decide to attend PAX East. Yet another distraction to endure. We briefly thought about showing one of the other flashbacks, but sanity prevailed and we took the GDC demo and continued to polish and harden it.
The GDC demos to the press were "guided" demos. They had the option of playing, but we were always by their side to help and warn. The PAX demos would be unguided. Players were left to their own devices to poke and prod wherever they pleased. We needed to make sure we plugged every hole. We needed to test as if it was a shipping project and this takes a lot more time. To the ThimbleTesting team's credit, no major bugs were found at PAX East, everything was rock solid.
Jenn's job on the project is the programming the Hotel and Franklin, since we weren't going to be showing her area, she was free to help set up our PAX booth and all the merch. As the pictures show, she did an amazing job with a very small budget. One of the smartest things she did was put two stools at each station so a friend could play. Adventure game are always better when shared.
Getting the demo ready for PAX took another 3 weeks out of our schedule. Again, none of it was wasted work, it was just distracting and felt like we weren't making any real forward progress on the game.
I was getting severely distracted with managing the project and struggled to find time to do programming that wasn't just fire-fighting bugs.
The role I've always wished we had was a producer, someone to manage all the schedules and sprints, and keeping an eye on the big picture while we forge ahead with puzzles and art. It's a role I'd been taking on and the burden was starting to show. I was spending more time working on spreadsheets then doing programming, design and writing.
After some budget analysis, we decided to bring on Chase Martin as our producer, a role I wish we could have filled months ago. We didn't really have the budget for a producer for the whole project, but coming on at this stage was doable. With Chase on board, I'm hoping to have more time to focus on my other three jobs. Hopefully it will make things better for the rest of the team as well.
The UI was starting to bug me. I love the C64 font, but seeing it on the screen really pigeonholed the game as a retro-game, despite it being much more than that. As more and more people looked at the game, we realized the font was becoming a limiting factor, much more than the verb UI.
Our goal has always been to capture the charm of the classic adventure games, but also to introduce them to a new audience without compromising what the game was. The C64 font was a hard thing for people to get around.
In the weeks leading up to PAX we tried a lot of fonts. Our tester builds had a new font every few days and none of them were clicking. We tried nice smooth truetype fonts, we tried pixelated truetype fonts, we tried crazy bitmap fonts and boring fonts. Nothing felt right. With the help of an outside designer we came across the font you see below. It is a hand-drawn pixel font. It's sharp and clear and you can see the pixels. It's a font we could have used back at Lucasfilm and it felt right.
The plan was, and still is, to retain the C64 font and allow players to switch with the press of a key. If you like it better, then please play with it. It's a good font and we don't treat it as second class.
For the opening scene of the demo, we wanted a full screen shot of the agents at the body. Mark extended the screen just around the body to full screen. Once the opening was over, we'd switch on the verbs and be back to the black cropped verbs.
While installing the new font, I accidentally left some code commented out and the verbs were drawn over the background, without a black background, and it was stunning to see. The game had a whole different feel. It took me around 10 seconds of walking around to realize that the whole game needed to be like this.
We had one week until PAX and our demo included over 15 rooms, all of which needed to be extended.
Oh... and Mark was gone for two weeks!
Octavi to the rescue as he took on the job of extending all the rooms in record time. Nothing playable happens below the interface, so it didn't need to be that interesting, and it actually wants to be uninteresting.
It really changes the feeling of the game significantly, but still retains the charm of the verb UI, something I didn't want to lose.
For those of you who want a more retro experience, not only can you switch back to the C64 font, you can also turn on the black verb background. But that's not all! Don't order yet! You can set the opacity of the black background to anything you want.
PAX went well. We had a great booth (thanks Jenn) and had hundreds of people playing the game. Almost everyone who sat down to play finished the 20+ minute demo and no one rage quit. I'll take this all as a good sign.
Where We're Going
Now it's back to work. We don't have any shows coming up in the next few months, so we can get back to focusing on the game.
As our projected summer release date got closer, I was starting to get really worried. Back in Sept, we had a lot of steam and it felt like we'd be done the following Sept or Oct, just a few months out from the Kickstarter date. But, as we hit April it just didn't look like that was feasible.
Well, not feasible unless we all went into crunch mode.
I don't like crunch mode. I've done a lot of crunch mode in my career and made people do crunch mode over the many years of running projects and it's just not something we want to do. We don't have an oppressive publisher looking over us and we have the flexibility to make the game anything we want (thank you backers!).
When Chase came on as producer, we did a complete relook of the schedule to see how much work we had left to do and how long it was going to take us. If we don't crunch, the workload puts us out to mid October, but we also have to go through the Microsoft cert process for Xbox, which can take one to two months. That would put us out in Nov or Dec and that isn't a time we want to launch. It's important that the AAA games have their day in the sun, and we didn't want to distract from that.
The other option was to start cutting. I feel good about the scope and size of the game, I don't want to cut it down just to make a ship date.
In the course of making a game, you make a lot of cuts for design reasons, and those are good cuts that make the game better, but when you cut for schedule and budget, you run the risk of cutting meat and not fat. That said, it's often hard to tell the difference, sometimes you think you're cutting meat, when in fact you're cutting fat and you're better off. It's often hard to tell the difference when you're in the kitchen.
But in the end, I decided I didn't want to hack large sections of the game away just to make an Oct date. We continue to make small cuts and refinements, but all those are to make the game better.
So we've made the decision to move the release of the game to January, mostly likely the middle to end of the month.
The budget is looking OK. With the addition of a few new and needed people, plus the extra time, things are getting tight, but we should still be good.
We raised a little bit of extra money through some angel investors for marketing and PR, two things that can be as important as making a good game. I feel good about using the Kickstarter money exclusively to build the game and the additional money to market it. It feels like a nice line, but it's also a little misleading. Marketing and PR is as much a part of building a successful game as music, art, and programming and they should be part of any budget.
In terms of the game, I think we're all feeling pretty good about it. It really feels like a good solid adventure game, just like we would have built back at Lucasfilm.
The one area that I'm worried about happens halfway through the first act. You unlock a large portion of the world and it's a great moment, filled with excitement and reward but players lose direction. It's a problem we'll probably solve with some good dialog and maybe a couple of new pinch points so there aren't too many new places to go. It's important to always give players focus. Player should always know what they need to be doing, but not always how they need do it. Being confused and lost is not a puzzle.
We started outside playtesting with testers culled from the readers of this blog. We did two people in Seattle and will now open it up to San Francisco and London. We have (literally) hundreds of people who signed up, so I don't know if we'll get to everyone.
As I've said many times on this very blog, doing playtesting is critical, but it can be time consuming. You have to organize people to come in, set up the location, spend several hours watching them, and on top of all that, you have to make sure the latest build works and is crash-bug free. It's a lot of prep and it's easy to keep putting it off, but resist the urge. Playtest! Playest! Playtest!
We have three bug testers on the project. Robert (lead tester) is full time and the other two are part time. We're looking to hire a fourth and that should round out the bug testing team until the end of the project. Our testers are amazing, some of the best I've worked with. Bug testing a game isn't fun and games. You're not being paid to play a game, you're being paid to break a game, then document it and figure out exactly how you broke it. It takes a special person to do this well, and they are gold when you find them.
What Scares Me
One thing that scares me at this point is the amount of work that needs to be done. It's a big game, but it needs to be. It's about the size of Monkey Island, 2 and to fulfill our promise of a "new classic adventure game", I feel it needs to be that size. I don't want to cut anything unless it makes the game better to do so.
At this point, it's about making smart decisions about the little things we can cut or rework to save time without compromising the game.
I actually enjoy that process. It's always been the fun part of a project for me. You need to make quick decisions about what is and isn't important. It really focuses you.
But it's also very stressful. It's one of the reasons I don't want to work crunch. Staying sharp can make all the difference.
Moving the game to January puts a lot of pressure on the budget. We had slop if anything went wrong, and although I wouldn't call moving the date "going wrong", it does eat up our budget slop. There is no more runway.
That worries me, but I feel like we have it under control. I don't think I've ever worked on a project that didn't feel like this towards the end.
The last thing is the amount of playtesting the end game will get. We've done a lot of testing of the early game, but we're still putting the end together and it's not in a outside player playable state, plus it's hard to jump new testers to the end of the game, so we need to pull groups back in for a 2nd or 3rd round.
Thank you to our backers and supporters for making all this possible.
- Ron