Blanket Fort Games Blog

We’re running a beta test at Cornell University, sponsored by the Digital Game Alliance at Cornell (the club devoted to the making and playing of video games).  Come join us at 1:00pm on Sunday, December 5 in Phillips 318 and play the latest beta version of Influence!

It’s looking like we won’t have the public beta ready before we all leave for thanksgiving, so I have uploaded a friends-and-family pre-beta release to the place where all of the previous evaluation versions are.  It’s set to expire in two weeks, but by that time we’ll have a real beta up.

Things to do before the real beta:

• Take another run at the tutorial, to make it simpler and more obvious how to proceed.
• Fix a couple of potential hangs if people disconnect from a multiplayer game at exactly the wrong moment.

Things to do before release:

• Improve scrolling and drawing when zoomed in.
• Improve load time by loading in the background behind the splash screens.
• Fix any other bugs found in the beta.

We’ve been working hard for the past couple of months to get to this point.  We have been certified by GameSpy Technology® so as soon as we fix one or two glaring bugs, we can have a semi-public beta of our game (to find all of the not-so-glaring bugs).  Expect an update early next week!

The current round of beta testing just ended, and if you’re looking fora new fix, you’ll have to wait a bit.  We’re integrating the 3rd party internet multiplayer tech, and so we’re under NDA until they approve a beta release.  It isn’t ready yet anyway, so y’all will just have to be patient.  We’ll be sure to let you know when we open up beta testing again.

After receiving quite a bit of feedback from various trusted beta testers, we are taking a hard look at not only the tutorial levels, but the entire single player experience.  The single player levels were designed originally to fit into a pattern based on the elements that they contained.  There was a section for levels with obstacles, with subsections for lines, triangles, and circles.  This division had three unfortunate consequences.

First, because our level selection screen was a tree-like structure, allowing players to explore the levels in whatever order they wanted, it was possible (even encouraged) to play all of the introductory levels first.  This meant that players would play all of the easiest, impossible-to-lose-so-we-know-you-understand-the-concept levels first, all at the same time.  Boring!

Second, because all of the levels with similar elements were grouped, players played them all in a row.  None of our elements were strong enough to design that many interesting levels without any palette cleansing in between.

Third, the number of levels was determined before any level design was attempted.  There were the same number of levels dedicated to the circle obstacles as there were to the super-ego.  Circle obstacles are not as interesting as the super-ego.  There were a surfeit of boring circle levels and a dearth of interesting super-ego levels.

Additionally, because we had the pattern, we made levels until we filled it, and then we stopped.  Only the most boring, unplayable, or gimmicky levels were cut for quality.

So what are we doing about this?  I’m glad you asked.  First, we’re adding elements to make the levels more involved, more challenging, more interesting, and more unique.  Second, we’re dropping the tree structure.  It just didn’t work.  We’re not sure what we’re going to replace it with yet.  Our plan is to make some really good levels first, and see if a structure for choosing which to play suggests itself.

Ha ha.  Not really.  The beta versions of Influence that we have released up until now just ran out.  If you downloaded your copy from our server, you will be able to find an updated version with a later expiration date in the same place later today.  If you got your version on CD, please send me an e-mail, and I’ll give you the download location and some login credentials so you can participate in the extended beta.

In either case, please please give us feedback.  What works, what doesn’t work, is it fun?

Thanks for participating in our beta test, and thanks for playing.

Ben

One thing the focus on consistency has given us is increased graphical fidelity.  Have a look!

With our newfound independence and Chelsea’s departure to the west coast (Good Luck Chelsea!), we have taken a close look at what Influence needs to make it complete.  Our list falls into three columns.

First, we are trying to make all of the pieces fit together better, to make a more cohesive game.  A good example of this is yesterday’s change to the background.  We added some subtle rings in the background months ago, to disrupt the sameness of the flat color and add a little interesting motion.  Yesterday, we connected that to the game, so that when you are playing, the rings are left behind the ids, like ripples in a pond, and while they always took on the color of the ids in proportion to their population, now the ids drop rings of their own color, so the colored ripples follow the colored flocks.

Our second objective is to focus on the opening of the game.  We need to tighten up the first play experience.  Right now the tutorial is stiff, broken up, and generally not very good.  It is miles better than our disastrous first attempt, but it still needs a lot of improvement.  Over the next few weeks, we’ll be trying to change it into less of a tutorial and more of an introduction to the game, which you can learn how to play by playing through, as we tell you about colors, voices, and influence.

The third column is internet multipayer.  We found a technology partner at GDC, and we’ll be integrating their tech with our game (and reworking some of the game to better fit the tech) over the coming months.

Each of these directions is important and necessary to the success of the game, and we won’t be satisfied until they’re all up to snuff.

Since we started sending influence out for some limited playtesting, and started getting crash reports, I felt that the level of detail of the crash reports would improve if I removed the human element.  So I set up influence to e-mail me a stack trace whenever it crashes.  This will be disabled or made optional in the release version.

Today I began work on the single player campaign mode.  We designed quite a few new pieces to go into the single player levels, and I figured that the best way to set it up is to add them simultaneously to the level editor and the game.  That way, the other members of the team can play around with (and test) the new bits as soon as they are available.  I started with spawn points, so that levels can be designed without requiring a cloud of neutrals and a ring of adversaries.

Next on the list:  Escort waypoints, control flags, and then obstacles.