Sunday, December 03, 2017

A New Watch App

I upgraded from my 3 year old Series 0 Apple Watch to the new Series 3 with LTE and I love it. Having worn the original Watch for the last 3 years I've noticed huge improvements in speed and accuracy (especially Siri) with the new model. The heart rate sensor has improved, the battery life is amazing even when it gets a hammering from workouts and phone calls.

Anyway, I was so smitten I had to make something for it.

As I've mentioned before I'm developing all my games in Unity from now on. It makes sense as a) I can go multi-platform, b) there is so much more support/tutorials/help available for Unity and c) it makes 3D development so much easier.

But an app is different to a game, so I feel I'm technically not breaking my promise to myself to focus on Unity for game development :-)

So what is this app? It's called Skip and Save and comes out of my habit of giving up the occasional treat and making a mental note that I just saved money in order to buy something else at a later date. I know, it's a dodgy way of budgeting, but it works for me.

I figured this would be a good process to make into an app and it seemed perfect for the Watch where the ideal apps are very limited interactions that take no more than a few seconds.

Building the Watch App
The Watch app development environment in Xcode is really nice. And laying out an app is super easy - so much easier than laying out an iPhone app as I would later discover.

You know have the ability to mix SpriteKit and SceneKit with regular screens. I used this to add ticker tape particle effects when you successfully skip and save money, and cool firework effects when you reach your goal.

I wanted to use as much of the Watch as I could, so I added in Siri voice recognition, Scribble support and support for all the different complications that the Watch has. I also made sure that the Watch could communicate back and forth with the iPhone version.

An early Watch build with a bar to show progress,

An updated version with a ring to show progress.

The first version of the app used force touch to access the options to setup your savings targets and to reset progress or undo a skip. From my own experience with other Watch apps I found this to be very unintuitive and often hid away key features. So I tried having a paged app that the user could swipe to  other pages to see more commands. This didn't feel good in practice either.

The final look of the Watch app. A single screen.

I settled on having a single screen app with buttons on screen to show more options to set targets, undo skips and reset progress and another information button to quickly see your targets.

I also tried out a number of images for the buttons and settled on a mixture of my own designs and those from the Font Awesome 5 set.

Building the iPhone App
As much as I love my Watch, the reality is the market for Watch only apps is still fairly small, so I needed to make an iPhone companion app. So I began the fun task of using Xcode and interface builder to create the app.

Storyboards for Watch apps are awesome. The way stack panels work is great, and you can overlay controls and even add a SpriteKit/SceneKit scene under a view - which lets you create some great layouts that can do awesome stuff in a simple and intuitive way.

The iPhone version is a little more complicated. I couldn't mix SpriteKit and standard views so had to rely on Core Animation to create the effects I had previously done in SpriteKit. So the knowledge I had from making games sadly couldn't be applied which meant I spent time having to learn new systems.

I also found creating responsive views a lot more complicated than in WatchKit and struggled with simple things like making my stack view contents remain at a fixed size. I'd end up with a mess of constraints that I would have to delete and start over again. Luckily RayWenderlich.com released a Udemy version of their training videos which included sections covering the very stuff I was struggling with - like stack views and how to use keyboards with scroll views.

One thing I did do with the iPhone app was to use vector images instead of pngs. They worked out really well, so I went back and updated the Watch to use vectors as well.

The original iPhone app had a light design.
Original iPhone design


I asked for some feedback on social media and got some great advice. Brent Helsop, a local developer gave me some great feedback and so the final version changed to better match the visual style of the Watch version.

Final look iPhone version

Time Spent
I worked on the app part time - spending up to 2 hours on any given day on it. I had a fully working version on my Watch within 8 days.

All up it took me 34 days of part time development to build the app - 10 days for the Watch (2 of that learning how to make complications), 22 for the iPhone and 2 days for iTunes Connect setup, screen shots and asset creation.

Launching It
The next stage of the story is how do I market it and what do I charge for it?

I've devoured the complete series of Under the Radar podcasts for some ideas as well as read lots of articles from other app developers. Being a game developer I have very little idea of how the app world works and these resources have been invaluable.

I've decided to price the app as free with banner ads from AdMob. I did try out interstitial ads (which I know do way better than banners) but it just felt wrong. They work well in games, but in my app they were too intrusive. My thinking on going free is that the people who want to give up something to save money are probably not the sort of people to shell out real money on an app that they've never heard of before. So being free gives them a chance to give it a shot.

With the app done and price set to free I have to reach out to the folks who would love to use my app. That starts now as I wait for the app to launch early in January 2018.

The good news is I can get back to making games in Unity while I do my app marketing.

I'll let you all know when Skip and Save is live in the new year!

-Johnno


Monday, May 01, 2017

Learning Unity

I've been slowly wrapping my head around Unity. As I posted last month I took some time off developing Billy Carts to do a number of Unity tutorials and learn more about C# and the Unity way of doing things.

Things are progressing well and I'm starting to "get" Unity. My biggest problem at the moment is not knowing if they way I am solving a problem or implementing a system is the right way or the best way. The Unity tutorials, forums and google are certainly a help and I realise that I just have to plough forward and make stuff. I can always refactor my code later :-)

To help in my quest in getting a full understanding of Unity I've taken some advice from an old friend and am building a small game in Unity that is essentially a homage to an already existing game. Not quite a clone, but a game that has a very specific set of mechanics that I can emulate.

The reason for this is that I don't have to think about the game design and can focus on solving how to implement the mechanics and systems using Unity. Not having to think about the design of a mechanic but instead focus on implementing it is really accelerating my learning.

The game is simple and I've got the first stages of it up and running. I've given myself just over a month to complete it - which will probably end up being many times longer - but the goal is to take a game from start to finish so I can be exposed to as much Unity has to offer as I can.

I'll post some screen shots and more details soon!

- Johnno

Monday, April 10, 2017

A Unity Refresher

This week I decided to take some time off developing Billy Carts to focus on honing my Unity skills and learn more about the Unity editor. I have to admit I'm quite rusty from the time I spent away from Unity focusing on finishing up Ultra Dash, my Swift game.

And, despite having made quite a few games in my time, I'm not too proud to admit that I have a bit to learn still when it comes to Unity!!

So, rather than slogging on with Billy Carts and constantly googling "how do I do this in Unity again?" I've taken on the task of redoing some tutorials and going over some Unity resources.

As well as the excellent tutorials on the Unity site, I've also started redoing the tutorials from Ben Tristem's Learn to Code by Making Games course on Udemy.

Easy to follow video tutorials that will get you up and running in no time.

I'll also be revisiting his Game Physics - Introducing Gravitation & Rotation in Unity course to get the physics parts of my brain back into shape.

A great course to get your head around physics.

Redoing these tutorials is not only giving me a quick refresher in Unity but it's also exposing me to things I missed the first time around - like neat little short cuts and useful  keyboard commands.

My other source of reference for the next few days is the excellent Unity Games by Tutorials by the raywenderlich.com tutorial team. 
Unity Games by Tutorials book cover
These are great books to make sure you get off on the right foot.
Ray Wenderlich makes some great books that you can either buy as books or as PDFs and the cool thing is they actually update them as things change. The Unity Games book for example, has recently been updated to take into account new stuff in Unity 5.5. 

While I like to have physical books as reference next my computer to leaf through, the knowledge that the PDFs won't be out of date within a few months is a good consolation.

What are your favourite Unity learning resources? 

Any must have books or online courses that make learning Unity a breeze? 

Let me know!

-Johnno


Thursday, April 06, 2017

Back on Track!

So, Ultra Dash is done and dusted. Its approved for the App Store and is waiting for release on the 12th of April. I managed to do a few more minor tweaks in the last week, adding back the watch video functionality that I had removed previously.  I figured that launch time is when I'll get most traffic and it would be stupid of me to not have the option to let players watch videos to earn coins and unlock new characters and levels faster.
I added back the ability to watch video ads to earn coins!


I've also done an update to Snappy Word, adding a new game mode and fixing some crash bugs that I found using the crash reports that users submitted.
Sort mode is a new addition to Snappy Word

Now I'm getting used to programming in C# again after using Swift for the last month or two! I'm sure it won't take long to remember to keep adding semi-colons at the end of every line...

In much the same way that having time away from Ultra Dash helped me look at the game in a new light, having the break from Billy Carts has given me some new thoughts on the game.

The game play right now is very similar to Faily Brakes - you steer your cart as it careens down the hill avoiding stuff and collecting coins. It's a fun formula that has done extremely well for Spunge Games!
Billy Carts is currently a down hill runner
Given the power of Unity and the ability to rapidly prototype stuff and change things like camera angles and lighting it would be remiss of me to not take advantage of this and try out some new ideas.

First thing I'd like to try is changing the procedurally generated hill to a procedurally generated plane, so instead of just rolling down the hill, how would it feel to drive in any direction? This would change the end game parameters which are currently stay on the track and avoid stuff - and might make the game feel too easy - but it's something I'd like to do.

I did a prototype of something similar before Christmas and it scared the bejesus out of Pete and myself. The prototype relied on building an entire world (mocked up by grey cubes) which would have meant lots of unique art assets from Pete.

But in the meantime I've had some ideas on other ways to do it that wouldn't require building an entire unique world - so I'm going to see if this new approach works. It may well suck, but I have to try!

My approach moving forward with Billy Carts will be to take a more organic approach to trying out stuff, seeing what works and following those "fun" paths. I'm painfully aware of feature creep so I will attempt to stick to the core pillars of:

  1. Avoid stuff
  2. Collect coins
  3. Survive as long as you can to rack up the most points

So please join me us as we go on this slightly altered journey to seek out new fun.

Given my experience with beta testing using TestFlight for Ultra Dash and Snappy Word, I definitely want to start testing as soon as I have something playable. So if you're keen, send your Apple ID email to support AT redspritestudios DOT com.

- Johnno


Monday, March 27, 2017

The Making of Ultra Dash


Ultra Dash is a simple endless runner game for iOS that uses a one tap mechanic. Players collect coins while avoiding oncoming obstacles by tapping the screen to change lanes. The coins are used to unlock new characters and also act as a score to compare against friends. Ultra Dash is also playable as an iMessage game.

Avoid obstacles, collect coins. Simple!


The game is written in Swift 3 using SpriteKit. I absolutely love Swift. It's an elegant language that is simple to pick up yet packs some real power. I will miss it as I bunker down in Unity for my next project. But C# ain't too bad either :-)

My goal in making Ultra Dash was to make a simple arcade game and to do the art myself. I was warned by friends (and my daughter) that my art wasn't too hot and that I should hire an artist. But in the end, I figured that I'm making this game for myself and I really want to cover all aspects of development.

Play against friends in iMessage

Well, all aspects except for music. That was provided by Eric Skiff who has some amazing free music that he lets people use in their projects - check out his stuff here http://ericskiff.com

The journey from concept to App Store has been a long one. My initial commit to Bitbucket was back in June 4, 2015. Why the long development time for such a simple game?

Well, it's been a part-time endeavour that was never a top priority. It got shelved regularly for months at a time due to other commitments - and at one point I decided to call it quits and abandon the project altogether.

However, it was over the 2016 Christmas break that I decided to revisit the game. The time I had spent away from the project allowed me to look at it with fresh eyes - and it wasn't that bad. So I decided to focus on the core game and finish it for release.

Original mockup for the Start Screen

The original framework was inspired by Crossy Road and the innovative banner system that creators Matt Hall and Andy Sum created. Crossy Road did a great job in presenting the player with different options every time the player died - constantly surprising them with new choices such as rating the game, watching a video to earn coins, winning a prize, winning a character or giving them stats on how long until they can earn more coins.

After taking my hiatus and reviewing the game I decided not to follow that model. I ripped out the banner system, removed the coin machine and made the act of buying new characters all done from the start screen. And each character offers a different background and level layout.

How the start screen ended up

I also made a conscious decision to remove as much English text as I could, instead using emoji to tell communicate to the player.

When you die you get sad emojis... kind of like cartoon swear words!


As well as the framework I also simplified the control scheme and the core game loop.

The initial prototype was all about ducking under and over oncoming obstacles in one of 3 lanes while moving left/right to avoid those obstacles that were too tall to jump over.

 I experimented with a number of control schemes including left/right and jump buttons, twin touch (where you tap both finger to jump), release touch controls like the awesome SHREDD where you hold both sides to go straight in the center and release one side to move to that side.

Here are some of the option buttons I used to try out each of the different combinations of controls and lanes:
I tried a lot of control scheme
And different level layouts

And in the end I went with 2 lanes 1 tap!

In the end I opted for a simple one touch control system and 2 lanes so you can play the game single handed. I also removed the ducking/jumping element and the extra tall obstacles.

For monetisation I settled on showing an interstitial ad after every 5 deaths and a banner ad on the bottom of the screen. I used HeyZap to mediate my ads and added a Remove Ads button to turn off the banner and the interstitials.

To buy a new character you have to spend in game coins that you collect. I originally had the ability to watch a video to earn coins but decided to keep the game as simple as possible. At one point I considered adding an option to watch a video to unlock a character, a technique used by Folmer Kelly in some of his games.

Testing!
Beta testing with TestFlight has been a godsend. If you don't do external testing then you are missing out on a lot of great feedback. TestFlight and putting the game into the hands of folks has been invaluable. 

Wrapping Up
My goal in making this game was to satisfy an itch to program a game and make the art. It may not be the most beautiful game in the world but I have scratched that itch and can now happily move onto the world of 3D and Unity. I will miss Swift though... of course if the game does well I will still be coding in Swift to release updates. Oh, and I also have an Apple TV and MacOS version ready to go...

Lessons Learned
Okay, this isn't a new lesson for me, but it was just a reminder of something I've known for the last 20 plus years. Making games is hard work. Even the simplest, plainest games. They take effort. And there are so many things that make up a game beyond the core mechanic - especially in this day of mobile games and Steam. The little things that eat away at your time like achievements, ad networks, multiplayer coding, cloud storage - all this stuff takes time to research if you don't already know how to do it, time to implement and lots of time to test, fix bugs and retest.

So, to everyone out there that has shipped a game or is in the process of making a game - I have huge respect!

Hopes
One of my hopes for Ultra Dash is that it has the chance to find an audience. I am a bit worried that the art may impact on that chance. Anecdotally any screen shot of Ultra Dash I tweeted would get around 0 favourites while a beautiful 3D Billy Carts screen shots could get up to 20 favourites. People like beautiful games. 

This isn't a surprising revelation, but it's easy for people to forget this simple fact. 

I'm also keen to see how the removal of English text impacts on the game. Will it be downloaded in countries other than Australia, USA and UK? I'm also keen to see how the banner ads earn compared to the video - I'll be looking at all of this data to help influence how I approach ads and UI for Billy Carts.

Well, that's my 2 cents on making Ultra Dash - it launches on April 12th so I hope you'll download it and check it out.

To wrap up here's a list of stuff I used to make the game.

Thanks for reading!
- Johnno

Development Environment
IDE: Xcode
Language: Swift 3 with SpriteKit

Tools
Art - Pixelmator and Affinity Designer to create the sprites and icon artwork
Sound -  Audacity
Icon Creator - Asset Catalog Creator

Music - Eric Skiff

Version Control - Source Tree and Bitbucket
Tasks - Trello to keep a list of To Do, Doing and Done tasks

Knowledge
Ray Wenderlich's site and books:



Platforms: iPhone, iPad, iPod Touch, Message
Coming Soon: MacOS and TVOS (if the game does okay).