Today’s article is a guest post from Squize over at the ever helpful Gaming Your Way. Squize takes an honest look at what went right and what went wrong in their most recent release, Invaders Must Die, an advergame for none other than kickass UK band The Prodigy, built in an amazingly short amount of time.

Invaders Must Die

To promote the new album by The Prodigy, “Invaders Must Die”, we were asked by Mousebreaker.com if we could create an on-brand viral game for them, albeit within a very aggressive time window. The brief was that we were to base the game on the album art and the gameplay was to be similar to the existing game, “Gun Run”, aside from that it was a blank canvas.

What went right

1. The Sound.

Let’s start with the obvious one.

The bands PR company originally wanted the entire album streaming, which is something I fought against right from the very start.

As a fan I could think of nothing worse than your first listen of a new album with a less than great bitrate, with a Flash game in the way. Listening to a truly great album for the first time is like a spiritual experience, it should be treated as such ( Even thinking about it now makes me shudder ).

We were sent the track, “Invaders Must Die” as a wav and I cut that into the title screen loop and the in-game one and dropped those in. Straight away it just worked, the title screen loop setting the tone perfectly and the in-game loop lasting long enough not to jar on repeated plays and yet not blowing the filesize out of the water.

In terms of the soundfx, the weapons were ones I’d originally put together for HeliAttack 3, which I’m not sure made it to the final mix. Others were pulled from existing games.

The collecting sound was from the next track taken from the album, “Omen”. We asked for the wav of it but were told we weren’t allowed it as it hadn’t been released yet ( Even on an official game you don’t get the whole world ). This lead me to find a cut down sample of it on a ring tone site where I could get the keyboard sample I wanted ( I did actually want the different notes that play around two thirds of the way through the full track, but had to settle for just the one note. In hindsight that did save some code).

2. The Look & Feel

Based on the album artwork by Luke Insect, Ed the designer at ipcmedia drew up some look and feel images for sign off.

First draft mock-up, featuring “Mini Keith”

As you can see they evolved a lot since the first concept ( As most games do, it’s very rare that anyone will nail it first time ).

When we were discussing the direction internally I used the analogy of

“The game should look like a flyer / poster you find on the ground. It’s got to look dirty and nasty, the sort of thing you’d only pick up between your finger and thumb right at the very edge, and hold it as far away from your face as you can whilst still being able to look at. It needs to look like it’ll cut you, and it will”.

It was key that we matched The Prodigy’s image, that aggressive angry burn down your fucking street vibe, every things shit and you can either accept it, or fight against it. The game had to look dangerous, and I think with the added effects and over the top distortion transition effect we achieved that well.

3. The Game Design Document

Although we were working to an existing game style, the brief literally was what was outlined in the opening paragraph.

We took the decision early on to produce a GDD in our own format, both as a bible for ourselves, and to get sign off from all the shareholders ( ie, Mousebreaker / ipcmedia, the record labels PR company and the band themselves ).

Original ground based baddie.

Although this took some time it was produced whilst we were waiting for sign-off for the look & feel and so had no impact on the actual development time, and a GDD is always a worth while investment.

4. The team

This may sound like a trite comment, even a kiss arse one, but it’s very valid and impacts on all projects. As developers we’re the first to bitch when we’ve got a project manager not doing their job. Both Kathryn at ipc and Rich at Mousebreaker were both excellent throughout the development, both being very supportive as they were fully aware of what we were trying to achieve and the timescale in which that was going to have to happen.

As a developer you want to be shielded from client demands as much as possible ( Too honest ? Maybe, but true ) and have someone fight your corner on the amends, without you having to do it yourself all the way through. Kathryn did that perfectly, her juggling of everyone involved made the whole project that bit smoother.

What went wrong ?

1. The Asset pipeline.

As it was ipc’s first “in-house” game there was a learning curve for everyone involved. When we were first approached to work on the project we were asked if we would be using our own designer ( Ha, the luxury of having your own. That’s like having a toilet with an automatic seat, yeah you want one, but who can afford such a thing ? ), but the budget didn’t allow for us getting someone in for it, so they turned to an in-house designer.

With the time against us I went with the decision that it would be quicker and easier to tweak the assets myself rather than explain exactly what would be needed.

This led to all the assets being in psd’s which were unfortunately 30% larger than the actual game. After some initial hiccups we got the asset swapping a bit more refined, although with hindsight I should have laid down exactly what I wanted more, rather than just trying to absorb that into my workflow. It’s more than enough work to skin a game up without the added step of having to convert the assets into something usable.

(Aside from that, the design was quite fluid, so instead of me working drone like to produce an exact copy of what is in the psd as is common in a lot of design agencies, I had to take a lot more on board in terms of layout and even copy writing).

There is also a world of difference between Photoshop design and designing for a Flash game. Some of the assets, such as the screen crumple effect, were created by using blend modes in the PSD ( Screen in this case ). Zero processor cost in an image document, very different in a game, which caused me some concerns about the potential impact on the performance, but luckily it seemed to not harm the game.

I think there were a lot of learning points for everyone involved in the art side.

2. Mini Keith.

A lot of time was spent waiting for sign off on the main character design, “Mini Keith” as he became known.

This was a cause for some contention for me personally, as I hated everything about the idea. To lift directly from the GDD,

“The avatar will need to be abstract as we don’t want the player attaching themselves too closely to any particular image (Aside from the Ant logo the band don’t have an external persona as such), in the same way a first person shooter encourages detachment for the player (We can all relate to Lara Croft or Mario, but the Marine in Doom is no one, an everyman).”

There is a really interesting email dialogue about it, which for the sake of some sort of professionalism on my behalf I’m not going to share, but it ended with me sending a dismayed “oh” as my final word on the subject.

Thankfully the band saw it the same way as me, and turned it down flat. This had the knock on effect that we had no player sprite a couple of days before the game was due to be finished, which lead to a real rush and the next thing that went wrong…

3. Animation

Due to it being a learning process for the team at ipc, a lot of effort was spent trying to nail everything down so there were no nasty surprises for anyone ( In saying that, there are always last minute amends. Even as the game was due to go live the client requested mochi-bot tracking in there, which caused other problems ) but there was one which almost de-railed the entire project.

Mock up of replacement player sprite.

With the delay on sign-off for the main character we were left with two working days to animate the above sprite ( Which was based on security camera images on the back of the album artwork ). Due to slightly crossed wires it was assumed at ipc’s end that I’d be animating it, I think due to Flash being seen as an animation package, so I’d be able to do something clever with code.

My code isn’t that clever unfortunately, so it was left to Rich at Mousebreaker to rush through the anims ( The first being a large chunky ED209 style robot, only to be replaced around 5pm on the final Friday before the gold date on the Monday, with the image that is in the game now ).

Our one regret is not having the chance to animate the player targeting properly. I did spend some time trying to alter the angle of the players weapon, but it would have meant animating the arms position to align it correctly, and I just didn’t have the time ( Nor the time to have the players head following the targeting cursor )

If there’s any one thing in the game that looks cheap and spoils it, for me it’s that.

4. Timing

Every developer / designer will always tell you there’s never enough time for a project. In terms of actual development, this game was started on Thurs 22nd of Jan, and finished on the 2nd of Feb, with minor amends after that ready for the launch on the 4th.

That’s 7 working days, although I had a meeting on the 23rd of Jan, so that day was written off. To try and make a great game in such a short time span could either be considered conceited, or considered the norm for adver-gaming. It’s really not ideal.

On a personal level it mean working a lot of late nights / early mornings and writing off both weekends. In terms of the game, it meant we only just managed to get everything in there that was planned ( In saying that, there was a long delay in confirming the wallpapers, and they were only agreed on the last Friday, and the code not added until the Sunday evening, the very last part of the game to be done ).

It meant that we didn’t have chance to play test it for long enough, not in terms of bug testing, but in fun and balance.

Because of that, the difficulty curve is flawed.

In conclusion

Is the game a success or a failure ? In terms of it being a game to promote the album and keeping true to the spirit of the band, then I think it’s a success. What it does, it does well. It’s a great viral adver-game, even if as a game it is just ok ( I really don’t see any contradiction there ).

All three shareholders are happy with it, so now it’s just down to the public to enjoy it, and it’s job done.

On a personal level, it’s a great IP for gamingyourway to get, for me it signals us taking a step up, that bit closer to where we want to be, and a vindication of a lot of hard work for the past couple of years. Hearing that “Liam LOVES it”, knowing the whole band have played it and really dig it, well shit, that’s what it’s all about isn’t it.

There have been a lot of things to take away from this project, the constant of trying to do so much in such a short space of time being the lead. It’s a catch 22, do you turn down a chance to work with such great IP for fear of not being able to do it justice, or do you take it on and run the risk of selling both the project, and yourself, short.

In terms of development, I should have set the time aside to spend going over the asset requirements rather than at times taking the martyr approach, as that would have helped everyone.

Every project teaches you something new, it’s up to you to actually remember what for next time.