Some quick thoughts on Zelda Breath of the Wild

I tweeted the above thoughts on Breath of The Wild after watching waaaaay too many hours of the Nintendo E3 Live Stream. Someone on Facebook asked me to elaborate, so here are my thoughts with a bit more unpacking. 

So Looking Glass Studios were the originators of the first person simulation genre with games like the original System Shock, Thief I & II, Ultima Underworld, etc.. A lot of the games that continued in that vein were made by ex-Looking Glass folks, the first Deus Ex, Bioshock, Dishonoured, etc. These games tend to try to simulate a lot of interesting interactions and prize player agency in creating their own solutions to problems. Enemies also tend to have more complex AI, with some simple emotional states and awareness of the world. However they can suffer from sticking too closely to reality, simulating things that just aren’t interesting and doing a lot of things but doing none of them particularly well (EG stealth in the original Deus Ex was an option but was terrible in practice).

Nintendo design tends toward elegant simplicity, they’ll build a game around a small number of mechanics and execute them really well, exploring a lot of the possible uses for them and making sure everything works well. Nothing is in there that doesn’t need to be. If there is a lot of content it usually doesn’t interact in as many ways as it would in a Looking Glass style game. Nintendo tend to focus on gameplay first rather than worrying about realism or coherent worlds. Puzzles often have a single solution and will sometimes block the player from using their own creative solutions (invisible walls in late stage Majora’s Mask dungeons I’m looking at you!). 

Both of these philosophies have strengths and weaknesses but what’s great about the mix in the new Zelda (based on what I’ve seen) is that they’re working together to make something with the strengths of both while removing the weaknesses. There’s some super robust simulation stuff going on but none of it feels needless or out of place. It’s not overly concerned with making these simulations realistic, so they can exaggerate them to create better, more interesting gameplay (EG spicy food protects you from the cold, carrying a metal sword increases your chance of being hit by lightning in a storm) and it has the level of gameplay polish and “game feel” you expect from Nintendo. 

A good example of a moment in the stream that felt pure Looking Glass style emergent gameplay was Link cutting some saplings to make sticks in a gale force wind. The wind carried the sticks into and through a campfire, setting them alight. The sticks then blew into a field of dry grass that caught fire and quickly spread, alerting a nearby encampment of enemies.

Zelda has flirted with emergent stuff for a long time but it’s never been this coherent and all encompassing, tending towards small one off touches that don’t interconnect above a certain level.

Fun side note. Looking Glass’s best, or at least most elegant games were probably Thief I & II, for which the designers have mentioned being highly influenced by Nintendo’s approach to design. 
__________________________________________________________
In other news we’re Kickstarting a new game “Russian Subway Dogs” soon!

Playing with risk

Here's an interesting tumblr post on how both Shovel Knight and They Bleed Pixels add elements of risk and reward to checkpoints. I wasn’t aware of Shovel Knights’ checkpoint destruction for cash system till reading this. Great to see other game devs experimenting with systems that integrate checkpoints and saving into the gameplay in a meaningful way. There’s a lot of unexplored territory there.

By the way, our friend Mathew Kumar who originally suggested the They Bleed Pixels checkpoint system has an upcoming game called Knight & Damsel in the works folks may want to check out. He also designed a particularly devious They Bleed Pixels guest level coming in the next update.

Playing With Risk

A Short Dance About Architecture (and Action Games)

A short game design microtalk on performance platformers and game controls I gave at the Tokyo Indie Underground event at 8bitcafe in Shinjuku (2013/09/18). Worth checking out if your interested in some of the thinking behind They Bleed Pixels.

Into my second week in Tokyo now and lots to write about including attending and exhibiting at Tokyo Game Show’s first indie game event (including a super confusing stage show I got to be a part of).  More soon!

Development Video #9 - CCTV Cameras

A bit of a longer video, it's about the CCTV Cameras in the game, and since that connects to some of the themes and inspiration behind the game I took the opportunity to talk a bit about them.  Each of the last few videos has been a bit of a different take on what the GG:SoR dev vid's are about, I'm probably going to continue along that vain.  Making a game (any game) involves so many different processes, activities and ways of thinking and I hope that by taking this approach I can build up a broad and detailed view of what's going into making this one.

All footage is from our prototype, the final game will look more like these: 1, 2, 3.

Thanks to Emma Byrne for letting me use her fantastic photography of CCTV cameras  in this video. You can see all the photos and read the accompanying essay by Cory Doctorow  in her photo essay "Snitchtown" at http://snitchbook.notlong.com/

The Torontoist article on CCTV cameras that I briefly showed in the video is here.  They do a great job of covering local public space issues.

The photos of plants were from my first time guerrilla gardening that I posted about a few months ago.

I mentioned that Shenzhen China is being used as a testbed for networked cameras employing facial recognition and software that alerts the police when an unusual number of people gather in one place.  I'm unsure of how much of that has actually been rolled out so far but  if you're interested in more info, here's a pretty thorough article on China's Golden Shield.

On the other hand if you're all like 'Whatever, F*@k privacy!', and you'd rather glory in the abundance of ubiquitous surveillance cameras available for your voyeuristic pleasure SurveillanceSaver lets you see through over a 1000 unprotected CCTV cameras around the world. You may not want to run it all the time, but I found it fascinating to view so many random places around the world in real time.... also kind of creepy when you realize how many of these things are listed on google without any sort of password protection.

Development Video #7 – Game Balance and Tweaking the Screamers

Because of TOJam and the like it's been awhile since the last proper Development Video.  This one is longer and more in depth then previous videos and covers some of the balancing and tweaking we've done to the perennial screamers in detail. 

GG:SoR's game design is largely systems centred, it's about complex interrelating systems rather then pre-scripted solutions.  So getting all the elements to balance and work correctly in concert as we add more plants and AI behaviour is our main challenge now that the basic gameplay is in.  We need to make sure that solving levels is largely about coming up with solutions based on how the various elements, cops, citizens and plants interact and making those  interactions as transparent to the player as possible.

In specific terms there's just no way for us to predict what a player is trying to do with a screamer and which way they want to scare a citizen, so we need to make the screamer's behaviour predictable enough that the player knows how to place them to get the result they want without them having to think about nitty gritty details we may have going on under the hood.  If as the designers we do a bad job, and a seemingly perfectly placed screamer result in the citizen running off in the wrong direction for no clear reason the player will feel frustrated and betrayed.

As I mentioned in the video, I couldn't fit every aspect of tweaking the screamers into it.  For those who are interested here are some of the other issues we faced:

Along with coming up with behaviour that would work we also had to balance individual variables such as the size of the trigger zone, the length of time it takes after a perennial is triggered before it screams and how long it remains dormant before turning red again.  These factors are all interrelated and, as we discovered, none of them are trivial.

For instance one build we had with a slightly shorter dormant time resulted in a cop getting stuck, endlessly revisiting the same screamer surrounded by slowgrass.  The slow movement speed meant he was still trying to escape its trigger zone when it turned back to red causing it to scream and attract him back in an endless loop.

One behaviour we added that I left out of the video was a system Andrew suggested where the screamer's trigger zone would slowly expand if left dormant for too long, ensuring it would eventually scream after a few minutes.  However that messed with our ability to direct citizens since the delay before screaming was tuned specifically to how quickly a citizen could walk past the flower after hitting the trigger zone. Our solution was to delay the scream by the same factor that we expanded the trigger zone. Bigger zone.. longer delay and the civilian would still be scared foraward past the plant.

Hope you find this longer more detailed Dev Video interesting.  If you did let us know and we'll do more, if you hated it and prefer them short and simple post a comment and we'll avoid them in the future.

Oh also we're switching to a one video a month schedule for the next little while so I can concentrate on hitting some production deadlines.  But I'll be sure to keep posting interesting stuff on the blog in the meantime!

New Build of the Game and My First Time Doing Real Life Guerrilla Gardening!

I was up late last night with Andrew prepping a stand alone copy of the GG:SoR prototype and submitting  it to IndieCade for consideration in this years show (I also submitted Cephalopods). Then I was up relatively early today for my first ever real life guerrilla gardening experience!  Sort of practising 'method game design'.  Here's a gallery of photos:

I love that the little corner plot my sister Mara and I planted with local plants ended up with some kid's Pikachu painting on its 'please water me' sign. A video game related sign really does seems appropriate.  The whole thing was organized by the awesome folks at the Toronto Public Space Committee who are involved in a whole bunch of community projects in the city and  as it turns out the spot they'd chosen to garden was conveniently at the end of my street! It was a lot of fun and a lot less hard work then I was expecting with so many people helping out, I'm hoping to get out a few more times over the spring and summer and do some more. Maybe also make some seedbombs if I have a chance.  EDIT: there's a posting with more photos up on the Toronto Guerrilla Gardeners Blog.

For those who are interested in what the new version of the game we sent IndieCade looks like here's an in game screen-shot:

View fullsize

Screen-shot from current prototype (click to enlarge)

As you can see the new GUI I talked about last post is in.  It not only looks better but also makes it much clearer what flower is selected, how many seeds you have and whether or not it's a plant type you can throw over an area (left corner),  or have to dig in a single spot to plant (right corner). I'm looking forward to showing this version to some new testers and seeing how much this helps them learn the basic game play. We also added more of the sounds I recorded before GDC in.  Unfortunately some of it isn't working for me... we have this idea that smell-so-goods smell like tasty food to civilians, however most of AI "barks" I wrote to convey that idea just react to the smell ("Mmm... who's baking pancakes?") without making it super explicit that they're reacting to a smell ("Mmmm... it smells like fresh pancakes!").   Unfortunately the sprites are too small to clearly convey that they're smelling the flower's pollen so it just seems like they have a random love of various foodstuffs.  I'm going to add a visual 'smell' emote to help, but now I  know that I need to be overly explicit when writing "barks" like these for the final version of the game.  It would probably also help to add a custom "bark" when they arrive at the flower, something like "huh, where are my pancakes!?".   On the flip side the audio for radioing in a tree drop worked like a charm:

Because the prototype is a pretty stripped down and doesn't have any sort of scripting system for tutorials, the current playable version uses a series of short levels with one page tutorials in between.  Clumsy and a bit text heavy but I'm hoping they get across the info someone new to the game needs to play.  I'm pretty happy with how they look given the constraints.

View fullsize

An example tutorial page (click to enlarge)

Development Video #6 - Influencing Citizens.

Here's the second video for the month.  Continuing from last video I'm concentrating on citizens.  This video shows how to use perennial-screamers and the new 'smell-so-goods' flower to direct citizens toward flower arrangements you've created. You'll also hear some very early sound implementation work we've done.

The level I'm playing in this video is actually a slightly modified version of one of the tutorial levels in the demo I was showing people at GDC.

As always, this video is of our prototype and has a lot of quick rough art, if you’re new to the blog you might want to check out my previous post, previewing what the final game should look like.

Development Video #5 - Happy Citizens.

As predicted this month's first video is a little late due to my much needed post GDC vacation.  I'll be putting a second one up next weekend and then we'll be back on track. The last few videos have concentrated on the stealth aspect of the game: tricking and evading the cops.  This video switches focus to the citizens and making the best use of ornamental flowers to cheer them up so they'll rebel against General Bauhaus's fascist government.

As always, this video is of our prototype and has a lot of quick rough art, if you’re new to the blog you might want to check out my previous post, previewing what the final game should look like.

Development Video #4 - Tactical Tree Drops!

This month's second video is all about 'tactical tree drops'  one of the first ideas I had when scribbling down the concept for the game in one of my sketchbooks.  I think this feature really captures the over the top take on guerrilla gardening I want the game to have. Needless to say I've been looking forward to being able to demo it for awhile.

So yes, you can radio in air support to drop a tree nearby, perfect for hiding from the cops.

This video is of our prototype and has a lot of quick rough art, if you're new to the blog you might want to check out my previous post, previewing what the final game should look like.

The next video may be a little late since I'm going to the Game Developers Conference next week. On that note, if you're an online distributor, publisher, press or the like and would like to meet, send me an email.  I'd be happy to demo the prototype for you!  (Same goes for fellow indie game folks of course!)

Development Video #3 - Shorter, Faster... Better?

Here's the first of this month's development videos.  It focuses  on a new plant we've added called the 'perennial-screamer', for reasons that will become obvious when you watch the video.

As I mentioned last post we're trying something new, cutting down the length of the  individual videos and posting two development videos a month instead of one.  I also experimented with zooming in and cutting the game footage so that it's hopefully a bit clearer what's going on.   So while it looks pretty chunky and old school the game actually runs at 1024x768.

If you want to see a sharp version of  the new art in this video, I'd recommend checking out the previously mentioned  T.O.Jam posting about the game.  It shows off the over the top statue of General Bauhaus, a perennial-screamer and a useful flower who's identity has yet to be revealed, though fans of obscure flora may notice a certain resemblance to a real life flower.

Development Video #2

A little late but here is this months development video!  Andrew has been busy getting features into the prototype and stomping out bugs so this time I recorded it solo.  Last month all we could do with the prototype was create a simple world with a few people wandering around and almost no interactive elements,  you'll  see we've come a long way and there's a good chunk of simple gameplay working now.

Because YouTube and pixel art aren't friends here's a still shot from the current version. Remember as we mentioned in the first video, we're keeping the art minimal for the prototype, so this doesn't represent what the final game will look like.

View fullsize