Defendy Rocket: Development Update #8

Before we start, here's this week's gameplay video which shows some footage from our current build. If you've been watching them all up until now, it's probably going to seem pretty similar to the last one, but we're at that point we're mostly just fixing up minor issues and tweaking performance:


In the last development update, we mentioned we'd been planning ahead and had come up with a simple schedule for ourselves. Here's how we've been getting along with that schedule:

6th August: Initial release of BETA #2. Medium difficulty implemented and first pass balanced. Hard/Endless modes not ready for testing at this time.  

Complete! We rolled out the release of our second BETA on time. This version was released to the same group of people that we rolled out the original BETA to, but we added a couple of additional people for good measure. Pretty much everything we intended to get into that release was there, minus a few minor things that we disabled while we continued to polish them. As planned, neither Hard nor Endless mode was in a state to ship at that time.

17th August:  Hard mode implemented. Systems for endless mode complete and endless mode implemented. All game assets, UI, audio, music etc complete and polished. Core game is finished and mostly balanced (balancing will continue until the end). 

This is a mixed bag. We did managed to implement hard mode for the 17th (although we didn't put in anything more than a few test enemy waves). Over the last three days we had Jack, a work experience student join us. We decided that helping to implement, balance and feedback on some of those hard waves would be an interesting exercise for him (plus he gets to see something he helped create in the final game). So Dave and Jack spent a day implementing a set of waves for hard mode and, outside of some minor feedback/balancing we need to do before we launch, it's pretty much done.

Endless mode is also now complete and have a whole bunch of test waves in place. All we need to do now is continue to add new waves and balance out exactly where they fit into the difficulty curve. And we'll probably be doing these sort of balance tweaks right up until release anyway.

Pretty much everything else in the game is done and in place too. We've implemented the last of the cosmetic items now (although we have a couple of minor polish items to deal with). We've also been through and cleared off most of our Trello board (we used Trello to track what we have left to do) of minor tasks and bugs. We also rolled out a lot of these improvements to the BETA testers.

There are still some bits and pieces left, but nothing that we're worried about completing in the next two weeks.

24 August: IOS code built, IOS achievements/leaderboards implemented.

We've been working on this over the last three days but without a huge amount of luck. Initially the plan was to get Unreal setup on the Mac, build the IOS executables and return to the PC version with them. But the editor refused to run Defendy Rocket on the Mac (starting the editor on its own or loading another project would work fine though). It eventually took two days of trial and error and then finally upgrading UE4 to 4.9 (preview 4) before it worked. And even then, despite being able to test the game on iPhone, creating a shipping build gets stuck (it doesn't fail, it just gets stuck). 

At least 4.9 seems like it's in a good state though and, ideally, we were hoping to switch to it before the end anyway, as it (potentially!) fixes an issue we have with our music getting out of sync.

At this point, I think to get IOS working, we're going to need to download and compile the engine ourselves (again!) and build the IOS executables remotely from the PC. Although that's a big task and is still not guaranteed to work.

We're still on track to have the game released at the end of September though, but it's going to be the next 5-10 days which is really going to determine if we make that deadline, or not.


Last Thursday, Epic kicked off a UE4 specific game jam (for those that aren't aware, game jams are (usually) short sprints to try and build a game in the given time based off of a provided theme). This jam ran from Thursday evening through to Monday morning. We decided it might be worth losing a day (Friday) from our schedule in order to participate in the Jam, just to keep our skills fresh. We waited to see what theme was announced and an Friday morning discussed what we could build.

The theme ended up being "Thrown in at the deep end". Obviously the first thing that comes to mind is "Swimming pool" or "Water", but we decided to go with an idea about someone being thrown into a job they weren't ready for. After about 30 minutes of discussion and sketching we came up with the following outline:

  • The player's been placed inside a factory, producing goods. It's their first day and they have no idea what they're doing, how anything works or what they're making.
  • The player can move around a small grid area, pick-up/drop objects and interact with them.
  • The factory has three different components: 1) Conveyors that bring in objects/components. 2) A work bench to construct objects and 3) An exit platform where you construct the final object.
  • If any one of the five conveyor belts fills up, the player loses and needs to start again.

Over the course of the day we slowly nailed the exact specifics of what the player was doing (they wouldn't be told any of this of course):

  • The conveyors would produce either objects or components. Objects basically be cubes and would be missing components (although the player would have to work out which ones) and the components would be nearly completely random. The player would have to organise these pieces and work out how to fit them together on the work bench. 
  • The work bench would allow the player to put an object on them and then manipulate that object by rotating it around. Components could be placed on the object and, when the component that the player was holding lined up with a missing component's location, they would lock together.
  • The exit platform would start with an object on it. The player needed to build a 2 x 2 grid of these boxes (again, they weren't told this). The initial objects three outer sides would be solid, while the three inner sides would either be male or female. The player would have to connect the corresponding, completed objects that they built on the work bench to this first object and, eventually, finish constructing the cube shape. The completed object would then leave the factory and they'd start again.

We put together a quick video showing most of progress we made:

Unfortunately, we didn't have the time to complete the project and submit it. Between not starting on Thursday evening, having important social events to go to at the weekend and preparing for Jack's arrival on Monday, it was kind of inevitable that we wouldn't get there. But none the less, it was a good exercise in prototyping and working within the constraints of the theme, so it was worth it none-the-less.

Posted on August 20, 2015 .