The Maker’s Eye (Downfall Dev Rundown)

Hey folks, Anzekay here, I’m one of NISEI’s Associate developers and Playtest Coordinator, and I am here to give a sort of brief peek through the cloudy windows of the development process that went into Downfall! Hunker down with a nice warm mug of something, and join me on this little venture of reliving the highs and lows of the past few months. Or, if you’re just here for more spoilers, you can scroll down a little bit and find them all right there.

The first question I should broach is just what the development process is, and what it is that our team does. Answering that is both very easy, and very difficult. Card development is, at its core, the process that takes cards from early concepts, to the fully-fledged and balanced results that are then passed on to the final stages of a set’s creation. In some ways it’s the glue that keeps a lot of the project together, since we have to keep everything neatly connected as different aspects of each card shift and change. Much of what happens to cards during development can affect the work of the other teams, and often we’ll be having conversations about one card or another with nearly every other team on any given day.

The difficult part of the answer is that the process by which we adjust, balance or even rework cards flows differently depending on exactly what a card needs. This is largely where playtesting comes in, the other extremely significant part of the process, and why I value our volunteer playtesters so highly. Every change needs to be tested, and feedback and data gathered and evaluated, before we’ll be content with how a card is shaping up. That test data and feedback from actually playing with the cards is vital. The Dev team could sit around and discuss cards all day, but we’d still never come up with any accurate gauges for the development of a set like we do from just a few days of playtesting.

Here’s where I should touch on the sheer logistical task that playtesting involves. We have to first assemble a playtester team from volunteers, and set up forums and other communication avenues. We need to make sure every playtester is able to access what they need to do testing. ZiNOS, my European counterpart in Dev, does immense work behind the scenes putting printable PDFs together for the testers, or feeding data into spreadsheets used by the jinteki.net team for the test server they so graciously set up for our use. Every part of this is vital, yet difficult to see from the outside.

Probably the single most important thing in any project like this is communication. I’m sure everyone who has ever done any sort of development project, tech or otherwise, can attest to that to some degree. The backbone of our team is without question divadus, our Lead Dev. His constant efforts to keep our design documents clean and up to date, talking to other department leads about any possible issue that might arise, and generally reigning in my often bizarre or outrageous ideas, are something I’ve come to value beyond anything else here.

As for the actual development cycle… Well, first we begin with an overall pass of the concepts the Design team have put together. Over the course of a week or so we’ll make comments, adjust a few numbers here and there and flag cards that need to be watched very closely. In rare cases we might send a card back to the Design team for a new version if we’re just immediately not confident that the card will stick. We’ll do that a few times later in development, as well, if a card isn’t turning out how we’d like.

Some cards actually change very little over the course of a set. Rezeki for example only had its influence tuned from 2 to 1 at the very end of testing. It’s hard to improve on simplicity, really. Cards like Calvin and Nanoetching received similar treatment, as did Tiered Subscription and Sandstone.

After that we begin playtesting, and start making new iterations to the card list every couple weeks or so based on a mixture of our own testing and analysis, playtest data, and general feedback from the playtesters themselves. That will loop several times, as we make each new iteration, and the set slowly crystallizes from raw concept to refined products. Sometimes we’ll have to rework a card entirely, or send it back to Design like I mentioned earlier. Sometimes we’ll have a request from the Creative team if we could shift the mechanical direction of a card to suit the theme they’ve decided on for a card.

Sometimes we’ll get really stuck on some cards, and revise them considerably with each iteration. Hyoubu Institute went through a dozen different concepts, some of which weren’t even based on the Hyoubu theme, before we settled on what we have. Rime began as a very humble piece of ice that had additions made to ensure the most basic concept of it was playable. “Baklan” Bochkin went through some wildly different designs and ended up as something quite different to his original incarnation. The new defensive tool for Weyland; Reduced Service was quite a difficult case to crack as well. Several designs proved too strong in some IDs but too weak in others, but we eventually settled on something that worked for most decks. By far the most significant card that had this sort of difficult development process; however, was Whistleblower.

Whistleblower started as a card that was a kind of one-use Film Critic plus Employee Strike; effectively blanking the text box of an accessed agenda. Clearly this is an extremely powerful card, especially for a Neutral card, and we tweaked various parts of its design for several iterations. The further we got in, the less convinced we became that the card was in a form we actually wanted to print, and several nights of fierce discussion and brainstorming ensued.

There was a brief moment where the card was a simple choice of gaining draw or credits, before we came upon an idea that was considerably more engaging. All this happened fairly late in development, as well, which made the process even tricker, but I am rather content with the result. What I really wanted was a card that proved powerful against the likes of Obokata Protocol or SDS Drone Deployment, but not one that would hit interesting, but not as terrifying, agendas like SSL Endorsement or Sting!. With some idea crunching we and both the Design and Rules teams put together a skill-testing card that worked perfectly, while still having a “loophole” we could design into if needed.

Around the start of the second half of the dev cycle (or a bit later) we’ll have the Rules team give the entire card list a thorough review, to help us work out kinks with how well cards can function or if there are legitimately game breaking issues with a particular part of one. They’ll be answering our questions throughout the entire cycle, but having this sort of review can help ensure that we’re able to get a decent playtest period with wording that is as close to the final product as possible.

Details are easily something you find yourself agonising over as well. As the flavour text on Nanoetching reminds us, even the smallest of mistakes or changes can create an entirely different result from the last. Blueberry Diesel flipped between allowing the Runner to peek at 1 card or 2 a few times before a late-night realisation led to my discovery of the middle ground that part of the card now inhabits. Flip Switch similarly felt cool but not quite there until one of my local players I was testing with suggested putting the old Disrupter ability onto it. Stargate was an obvious powerhouse of a card throughout most of testing, and the small addition of limiting it to once per turn ultimately made all the difference between a card that was oppressive to play against and a card that was an interesting challenge for both players.

Once we reach the end of the functional playtesting period, we lock in the general functionality of every card and spend a little bit more time testing and tweaking numbers and small details. Once that step is completed, we hand the whole set off to the Rules team to be tinkered with for wording consistency, and for overall editing!

Often there are particular philosophies of design that we approach different cards with during development. A card might be designated to be more of a top-down card, something where the original concept for a card came from its theme or name, rather than its mechanical design. Game Over was originally an entirely different card, that I personally spearheaded a rework of after the Creative team decided on keeping the card’s placeholder name. I wanted a card that felt truly fitting of its name, without… actually ending a game. Sorry, Adam players, but you’ll find solace in a shiny new breaker or two in Ashes that slots perfectly into an ID that starts with 3 virtual resources.

Loot Box is another card that was given adjustments intended to fit well with its theme, going from a simple piece of ICE to something with an appropriate joke built directly into its mechanics. The “Trio Connections” were built to evoke nostalgia and familiarity from the group up, after divadus had the ingenious idea of merging Noise and Wyldside into something so breathtakingly thematic I had to pinch myself.

Something noteworthy is that the approach to development on a game with such a powerful fusion of mechanics and theme means that to some degree every moment spent working on these cards we are suffused with the setting, world, and characters we’re dealing with. It can, at times, be all consuming. I might think about a single card during every free moment I have for several days in a row. Some of my best ideas or concepts have come during moments of travel; be it while waiting at an airport because my flight is delayed two hours, or just driving home from a day of classes at university one evening.

Fully Operational is an example of a card we altered to attempt to encourage a certain sort of deck archetype; something akin to the old asset heavy HB decks of old, where you’d make a few extra lightly iced servers to shove assets into. These days that’s a much more difficult affair, so we created a high-impact operation that combines with MirrorMorph’s tempo-positive asset package. Cold Site Server started out as a middling defensive upgrade, but we pushed its power a bunch to make it into a sort of dark horse card that can be surprisingly effective when played in unorthodox ways.

In a similar vein we have to give special attention to cards that are part of a ‘package’ so to speak. The Companions are an example of this, but Az McCaffrey and his console Masterwork (v37) are probably a more free-form demonstration of this. Az is all about hardware, something that is a bit unusual for a Criminal ID, and we had to slot some new hardware into the Ashes cycle (you’ll see more in Uprising!) that would work in tandem with this new paragon of Criminal deck types. Masterwork fits neatly into this, providing a hardware-based Criminal much needed draw and a run-based way to get click efficiency on your hardware installs. Tune in later this week for the Playtester Review and you’ll see the Shaper equivalent of this design area in the fascinating event called Khusyuk.

In the realm of complete card reworks (especially Game Over) we also had a unique opportunity to change up a few cards in the set to fit in with a sort of paired design idea. Originally Downfall only had two new cards that interacted with Bad Publicity, and both of them only got rid of it rather than produce it (Roughneck Repair Squad and NBN upgrade Increased Drop Rates). Partway through playtesting, I found myself looking at two cards that weren’t working out, and suggested we redesign them entirely to be illicit cards. That started us on the process towards creating Game Over and Trebuchet. The latter in particular nearly ended up being extraordinarily powerful due to the illicit downside, until we reigned it back in a little towards the end of playtesting.

Of course, at the end of everything I still find myself smarting from some harsh lessons learned, or regretful for how a card turned out. This is by no means something to be ashamed of; the learning process is a vital and healthy part of being involved in making games. We’re coming out of the creation of Downfall with every single department having set up important logistical assets, figuring out how to better manage communication within a global organisation, or just getting a feel for the pace and flow of working together on this marvelous project.

I have a few personal regrets, naturally, and I think airing them is a good reminder that it’s exceptionally difficult to put out a perfect product at the end of a development cycle. I wish I’d given some of NBN’s cards a bit more attention myself. I realised too late in development that I’d made a mistake when pushing a particular HB card design to Uprising, leaving more of a hole in MirrorMorph’s toolkit than I would’ve liked. I got very invested with working on the Companions, and ultimately had to take a step back from working on them directly myself after some concepts didn’t quite work out. As a result, I personally feel like I did the cards themselves a bit of a disservice. Thankfully the art for all these cards is so stellar I don’t feel gloomy when I look at them anymore!

But hey, this is our first set, and much like any game project I’ve worked on before, I’ve emerged upbeat and optimistic for my future work, and even more so for the work of my fellows in NISEI. The quality of work and quantity of energy that the entire team has put into this set has exceeded my own expectations, and that’s really what pushes me forward into the future.

I hope you’ll be along for the ride, too.

Author

  • Morgan "Anzekay" White

    Serving as Null Signal’s current Narrative Director, Morgan is a long-time writer and game designer from Perth, Australia, who has been in the organisation since its early days. Morgan currently works full time as a Narrative Designer in the gamedev industry, enjoys a good nap, baking, a good craft cider and watching sunsets during cloudy autumn days. They are also the near-full time Chief of Staff to Daeg, the cat and "official" mascot of Null Signal playtesting.