Unity 5 Reboot

Time to dust off this old blog.  Why?  Because I'm doing this game again.  In short, the HTML5 version went nowhere very quickly after my last blog post.  But the idea never died, it just lay dormant for a while.

It's back, better than ever.  HTML5, what was I thinking, Unity 5 now!

I actually blog this a few weeks into development, so I've had time to adjust my brain to the idea of building a real game, and I'm still not so scared off that I'm quitting, so it's time to blog about it again.

The aim of the game has actually changed very little from previous posts, fair to say it's virtually identical.  You get a player to lead for the ball, AI tries to stop you, your aim is to score goals.  I still think this alone can make a fun and interesting mobile game, so I'll be aiming for ios/android.  

On a macro level, I've added a few small ideas, and had a million ideas on the micro level.  But the bigger feature ideas include:

• Decision making is extended to where you want to lead, kicking to the right spot, kicking the goal.  AI will still do a lot of the heavy lifting.
• Can now defend against AI also
• Can play multiplayer against other humans
• 2D is now 3D, but more of a 2.5D game, as is intended to be a top down view gameplay (that in theory can be any view, perhaps even a user setting depending on their preference)

Now, I did just recently throw together a quick android app for the Olympics, a schedule app.  This went mostly the way I expected with just a couple of glitches:

• Don't try to write an amazing app in 2 weeks.  Amazing very quickly turned to glitchy, featureless and awkward looking
• Don't publish an app that isn't amazing.  The biggest problem with "glitchY', is that users are itchy to give a 1 star review.  Then the awkward look/featureless stopped people giving it a 5 star.  To you, it's your little baby that makes your heart shine to look into it's eyes, but to the rest of the world, it's just another smelly, screaming baby.  That analogy is a little harsh, but the overall point is, you have to be able to detach yourself from the app, and really criticise it, until it can stand up against your harshest criticisms.  Then you can publish.
• People don't think the same.  I talked to a bunch of people who installed this app, every one of them had a different way it should be to be a better app.  At first, this was taken a little to heart, "why can't you see that it's written in a really clever way."  But after the 3rd or 4th bit of "helpful" advice, I realised it's impossible to please everyone.  We do think of this in our day jobs, but it's always an eye opener when you get feedback that doesn't sound logical to you.
• UI is not my thing.  I know what I like, not necessarily how to do it well.  I had some ideas of how I wanted my app to look, it really didn't get there though.  It may be a case of more time and planning though, so it would be interesting to see how I go with a 2nd published app, which I intend to do after this game.
• Google does a pretty good job protecting the official apps.  I had to change a logo because it had olympic rings, name because it sounded like the official app.  I almost certainly won't be able to call my game AFL, as that is a brand, Aussie rules is the sport.  This kind of sucks, because nobody ever searches Aussie Rules, it's always AFL.
• And I sort of covered this in the first point, but polish takes time and planning.  Trying to force a product to a deadline has the exact effects you'd expect.  I had planned to have a learning AI server.  It would see what others "like you" had manually chosen as their favourite sports, and suggest it to you.  It would suggest Swimming for Australian fans, because we love our swimming.  It would suggest womens rugby if your team made the final, etc.  But in the end, the AI feature was completely ditched and you were on your own with choices of favourite sports.  It's a shame, because this was the part of the app I was most looking forward to writing.

All that said, I enjoyed writing the app, and it wasn't as flawed as I make it out to be, it just lacked some polish really.  I used it a lot, and thought it performed well for what I wanted.  

Now, the future is this Aussie Rules game.  I've got to allow myself a lot more time for this, I'm aiming at the start of the 2017 season.  Given there is a bit of time, I'm going to go about some proper planning this time.  So here I'll layout a fairly high level overview of the core goals:

• Goal 1: polish and publish.  I think I can knock up a playable prototype in around 2 months, so that leaves 4 months of polish.  This assumes of course I don't go down some other rabbit hole.  I'll give two examples of polish.  One would be AI.  Not only do I want the AI to be fairly bulletproof, I want people to walk away thinking about it.  So I want mindblowing AI.  You may say "how do you do mindblowing in such a simple game", the answer is of course, polish.  I'll get more into this in another blog, but I want polished AI.  Another example is user interactivity.  I don't want people left with too limited options, nor too many.  I basically want to have people play the game without too much learning curve annoyance.  A good example is having a tutorial, allowing the user to see it first time, but be able to replay it in case they skipped it the first time but now want to watch it.  Not critical, but nowhere near a high priority yet, but still would be part of a polished product.
• Goal 2: mentors.  I'll admit to be a bit of a hermit programmer.  My idea is always better than the group idea, at least to me.  A very experienced friend drummed it into me that I must do the mentor thing, it's not even worth doing a project without it.  This makes sense, but it's not how I wanted to operate.  However, I'm going to do it, because he is right, it's something I should do.  I haven't gotten the full details down on this, but I'm going to search fro someone to help me through this process, and plan to use the approach again if I make apps.
• Goal 3: contribute.  Similar to the mentor thing, not really my game.  But I've got a unity account, and should get active on forums.  I should also blog (tick), and tweet about it.  
• Goal 4: profit.  Of course, money?  Not really money, more like downloads, game playing.  I don't really know a realistic goal, so lets start with facts.  Fact 1, AFL official app on Android has 1 million downloads.  I'd imagine a similar number on IOS, and I'd imagine a percentage of people that have installed it on multiple devices that they don't even use anymore.  Either way, 1 million is an impressive figure.  Fact 2 and 3 are the official licensed AFL Live game is only 1 thousand downlods, and a non official game is 5 thousand.  Fact 1 is skewed, partially because it's a $10 game, which is expensive relative to other games, and it's also a huge download, which makes it impossible to install on a lot of devices, it rates only average overall (including a lot of cannot install/play reviews ) and finally, it's only been released recently.  That said, it has AFL in the title, is fully licensed and is a pretty good game, going by the ipad version a few years ago.  The 2nd game is more likely what I'm up against.  It's free with ads, doesn't have AFL in the title, and it's a small fun game.  My game aims to be a lot better, time will of course tell.  I'm going to aim at 2 thousands downloads in the 2017 season.  I'm also going to aim for at least 10 days of playing the game on average.  That may not sound like much of a goal, but that is already 200,000 game days, which will be near impossible if I can't deliver a decent game.  Actually, I'm also going to aim for at least a little forum posting about it (and not by me).   Money wise, it's probably a "for fun" project to be honest at this stage.  If there really is 200,000 game days (which I hope to accurately track using a web service, maybe), it would be silly not to expect a little advertising money may slip through.  Maybe enough to pay for that web service?  A very rough calculation says even if the pickup is 10x that amount, the advertising wouldn't be quit your day job sort of money either, so I might as well conceed this is a for pleasure project, not a for money.

OK, kick off time.  I'll be following this up with some more interesting blog posts (for me at least) about the actual coding challenges and ideas soon, which I hope may help some people more.