Protonaut gets Goombas

Protonaut has gotten universal praise on one thing, and one thing only: Shooting.

There isn't much to shoot in the game, and the feature is one almost stripped several times throughout development - but it's hard to scrap the one feature that makes people squeal with delight.

With some recent upgrades to the sound files, shooting is just getting more and more delicious - firing that cannon truly is an enjoyable experience. But I needed something a bit more to shoot at than the occasional Nitrogen atom.

Enter stage right: Goombas. Can you tell Greg has been busy for the last few weeks? :) My art isn't exactly befitting of the game :P

Goombas just walk left and right, and have pretty decent detection for obstacles and pits - and try to avoid them. I'm planning on make a jumper and a flyer of some sort eventually, but I'll see how this plays out in the next few builds. They kill you on touch and don't yet have any sort of projectile capacity.

Enemies might not end up in the final game. This production causes some fairly obvious problems, most obviously: the theme of the game is completely off with them. I'll see how things work out in terms of gameplay and see if I need to rework the whole thing.

It's too bad the Level Editor hasn't gotten more attention. I really feel like the level editor is a standalone work of art, and making things is incredibly easy and seamless - with tons of tools at your disposal. It's almost a direct correlary to my programming work not getting any notice, and the more artistic stuff getting the attention - nobody cares about the backend or the tools! It's like I made the best hammer in the world but I haven't found a carpenter to use it yet.

But such is the path of a programmer, I don't really have the ability to complain about that one. :)

Signposts: New feature on Probation

Protonaut seriously needs a tutorial overhaul, and I've been thinking about how to best accomplish this. I originally envisioned putting "signposts" in the game, with static text on them - then the tutorial would be one big level with a lot of help along the way.

Before I had a chance to put in the signposts, I did a few tests - some people couldn't beat my tutorial level! It was long, too difficult, and not very rewarding. An utter failure!

The signposts never made it in (instead opting for an "introduction" paragraph), and I broke the levels up into 10 easier (and more fulfilling) chunks. After launch, the complaints I get were - retrospectively - obvious: You now only get tutorial text at level launch, before it becomes relevant or obvious as to what it means. You cannot recall the tutorial text except by dying or restarting. A lot of folks took the tutorial text box as a "standard dialogue" and simply closed it.

One thing that really irks me is when players don't read, and close the box that says "PRESS Z TO JUMP - THIS IS CONFIGURABLE IN OPTIONS" then complain that they can't figure out how to jump or that Z is a stupid choice. Of course, I would do the same thing if it wasn't my game, so yes: You, Sir Kettle, are black.

So, here is my proposed solution: an in-game signpost. You can test it here, though the text isn't modifiable at this point (I need Greg to whip me up an interface to do so). I'm really interested to know what you think! Chime in with any thoughts or ideas. Here are some features:

• The intro-box will remain for two reasons: Need the Click-on-level-load to capture keyboard focus, and it gives level designers a place to put a little backstory up if they so desire it.
• The signpost text appears on touch only - thereby not "giving away" level elements or cluttering the screen.
• The signpost trigger box is sizeable any way the level designer wishes. As the text appears from the center of the box, this allows for some creative control on where text appears.
• The text fades out at a variable length, dependent on the 200WPM average that people read at.
• Linebreaks work and the resultant text box auto-resizes to fit.
• The signposts keep track of how often they have been touched, which can be recalled with the %COUNT% variable. This can eventually tie into a badge or victory condition (in the distant future) - allowing for laps counters or other more creative ventures.
• The signposts respect your keyboard control configuration with the %LEFT%, %JUMP%, %RESET%, etc. variables.
• The signpost trigger box is not yet art-ed up, art is not final :)

Again, try it out here, and let me know what you think!

Protonaut Beta Launch: A Retrospective

It's been nearly a month since launch, and it's time for a retrospective on Protonaut!

First up, I should explain away my absence for the last 30-odd days. I'd like to say I'm busy attending GDC and other business stuffs (which is partially true), but the actual reason has been crushing disappointment.

I really shouldn't be disappointed. My brain is telling me that everything is allright, and I know things will work out in the end - as I believe I do have a viable product here. The problem is, my heart doesn't handle disappointment well, and it's screwing up how my brain is responding. Just today I've managed to bundle up all the emotion and kick it out the door, and I'm finally ready to move forward.

Let's rewind a bit and explore what went wrong.

When I launched Protonaut, I knew I had a serious content problem. It's the classic chicken-and-the-egg scenario; I have a game that heavily relies on user-made-content to generate traffic, and contained very little user-made-content. All of my fancy level promotion and ranking tools are useless until I have mass traffic. I decided to launch an "open beta" of sorts, and get some traffic flowing.

Colin had a tremendous amount of luck with JayIsGames in the past, for his game Fantastic Contraption. The traffic was of the perfect demographic and seemed to fit with what I wanted my userbase to be, so I got a review up on there - then halted all media interaction immediately. I didn't want this to get out; it is still a beta, after all. The big release is yet to come!

And that's the traffic graph that pretty much overlays over the week I was at GDC. What's funny is I actually garnered more traffic than Fantastic Contraption did in it's first week, and I got more traffic from JayIsGames than FC did after the review went online. My entire problem lies in retention.

That there is a delicious pie chart of where people left my game for good, their IP address never to grace my website again. 45% of my visitors were lost completely before even loading a level. I then proceeded to lose around 20% of my traffic at each stop in the tutorial, to the point where less than 1% of total traffic finished the tutorial in it's entirety.

What's interesting, though, is that the average user spent over 10 minutes on the site, and loaded (or retried) levels on average 6 times each. A bit more digging shows me that quite a few people abandoned the tutorial partway through to get them to the user content. That's a lesson learned right there. Another lesson learned is that most of those people were restarting tutorial 1 over and over again - a sign that it's too hard, even in it's simplicity.

Sorry this one is unlabelled, but it is how many times each level was played - and essentially goes in order around the wheel - Tutorial 1, tutorial 2, tutorial 3, tutorial 4, etc...

Half the people that played #1 never got to #2. It could mean tutorial 1 is too hard, but it could also mean that the player has already decided "nope! this game isn't for me." A more engaging first-user-experience could correct this.

I'm a real slut when it comes to charts, graphs, and analytics. The average user only viewed/tried 6 levels, but sent me 15 hits to Analytics for clicking in various places on the screen. But the real blow to my pride was this graph right here - the one that everyone gets excited about, TOTAL CONVERSIONS:

Yes, that's right. I'm seeing a THREE PERCENT CONVERSION RATE! *balloons fall from ceiling, cue dance music*

Well, I did on that one day, anyway. If you average out the conversion rate since launch, it's at a still-respectable 0.4%.

Thing is, my traffic has been so low, that the numbers are heavily skewed. On top of that, I made 5 test purchases during development, and this graph is nothing but smoke and mirrors; I actually have made a grand total of ZERO sales of Protonaut after launch (except the one I paid my girlfriend to make).

Anyway, I'm putting my emotions aside now and gearing up to kick some ass. I'm going to rewrite the entire tutorial system and make the whole experience more engaging for a new user; I really want to ramp up my retention figures.

At least I'm getting "SSSL Unblocking" traffic from google!