Thank you, Guam!

A few things of interest:

• I started to color code the points by the time they registered, and then I thought I’d do a gif animation over the duration of the beta, and then I realized I’d already spent way too much time on this!
• I got a flurry of Japanese sign ups and tweets, mostly referencing this nice article on 4Gamer.net.  I should probably have the sign up page translated into Japanese and Korean…
• Are there really a bunch of sign ups from Arviat, NU, Canada, or is that a bug in the free IP geolocation database I used, or my code?  Also, Sao Tome?  Oh, wait, duh…Sao Tome is right near 0,0 lat/lon, so that’s probably just a bug!  Arviat is a mystery, though.
• I got the map image from here, but be warned, one of their “Mercator” projections isn’t.  I thought I had a math bug for an hour or so in the code I hacked up to render this.

Okay, back to work on the auto-updating and content packing!  Fun fun fun!

• At 17:05:12 PDT on 2011/5/11, somebody named Steven, whose IP address is from Florida, became the 5,000^th person to sign up for SpyParty beta access.  Thanks Steven, and everybody else too!  Oh, and an important note that people have been asking about:  everybody who signs up will get into the beta. I will talk more soon about how it’s going to be structured going forward, and I’ll provide a rough timeline.
• The page went public at 10:30 PDT on 2011/5/10, so that’s 30.5 hours, or about 164 sign ups per hour averaged over the entire time.  Here’s a graph of sign ups per quarter-hour:
  
• The beta announcement got a fair amount of press, which I really appreciate!  Here are some links, most with comments:
  SpyParty Beta Starting Soon, Here’s How to Sign Up – Stephen Totilo, Kotaku
  SpyParty ‘Early-Access Beta’ sign-ups begin, costs $15 when beta opens – Ben Gilbert, Joystiq
  Spy Party To Unwrap “Paid Beta” Access – Jim Rossignol, Rock, Paper, Shotgun
  SpyParty Goes the Minecraft Route, Launching Paid Beta – Patrick Klepek, GiantBomb
  Early-access beta sign-ups open for SpyParty – Jordan Devore, Destructoid
  SpyParty early beta sign-ups begin – Steve Watts, Shacknews
  I’ve Been Invited to a Party…. A Spy Party – David Liao, 4 Player Podcast
• There were a bunch of threads in various places, here are some major ones:
  Spy party beta sign up $15 – Reddit
  SpyParty Early-Access Beta Sign-up – NeoGAF
  SpyParty – You’ve done it Checker. I want to play this game. – Quarter to Three
  Spy Party: Read the Fucking Manual - Something Awful
  The SpyParty Facebook Page
• And, of course, Twitter.  See that bump right there in the middle of the signups?  That’s Markus Persson, a.k.a. Notch, waking up Europe with a very nice tweet to his 174,000 followers.  Markus was very helpful answering my questions while I made the decision to do the Early-Access Beta, and the tweet was an extra helpful icing on the cake.  So, of course, I asked him to tweet again mentioning @spyparty…
  
• One of the reasons I structured the Early-Access Beta this way, with the sign up first, was that I couldn’t really predict how many people would be interested, so I didn’t know how to scale the game server for the unknown load, and I figured it’d be easier to scale a simple sign up form than the entire game lobby and backend.  I spent a fair amount of time loadtesting and optimizing the webserver, and in the end it purred like a kitten the entire time, barely breaking a sweat.  One of the ways I prepared was using Amazon’s CloudFront Content Delivery Network for most of the static assets on www.spyparty.com.  Here’s that data…
  You can see the US and EU daylight hours pretty clearly here.  My total CloudFront cost from the past two days: $6.03
• Once I had the static assets on a CDN, I just had to worry about the dynamic page traffic on my server.  I’ll write a more technical breakdown of this on my personal website some time soon, but the loadavg on the server rarely went above 0.2, even when the Reddit article was on the front page of /r/gaming.  Here’s the bandwidth, with a 5 minute average (blue is outbound, green is inbound, it’s a log plot):
  You can clearly see Notch’s tweet here as well.  The big spike off to the right is the last load test I did before launching.  It’s interesting to see the quicker falloff of the tweet and EU daylight, compared to the fatter area when all the news sites were posting.  The double-hump in the first hit is when the Reddit post made it to the front page of /r/gaming.  Reddit drives an insane amount of traffic, even from subreddits, and even from the bottom of the front page of subreddits.
• Here’s a breakdown of the domains of the emails that registered with 10 or more registrations from that domain.  gmail.com clearly dominates amongst SpyParty beta testers!

  gmail.com	3216
  hotmail.com	522
  yahoo.com	273
  hotmail.co.uk	68
  googlemail.com	56
  comcast.net	36
  aol.com	35
  live.com	33
  msn.com	30
  me.com	24
  live.co.uk	14
  yahoo.co.uk	13
  mac.com	12
  web.de	10
  others	766

Overall, I’m very happy with how things have gone so far.  I honestly wasn’t sure what to expect, but the initial 1/2/7 day estimates I made were 2,000/3,000/5,000, and we basically blew through that in a day.  I’ll be interested to see how it evolves over time.

Now I need to stop checking my stats, and get to work on making this beta more than just an email sign up!

The Line Forms Here

Well, you all asked for it, so I’ve spent the past month trying to figure out exactly how to make it work, and I think I’ve finally got all the pieces figured out.¹

So, here we go:  I’m taking sign ups for the SpyParty Early-Access Beta, starting right now!

The sign up page is here.

I’m so excited!  This is going to be a lot of fun!

1. Famous last words…

Okay, here is the slightly infamous Four Page Instruction Manual for playing SpyParty at PAX.  I felt like I had to print this because of the one downside of the Depth-first, Accessibility-later development methodology I’m following.

My goal with printing up this document is to save my voice for actually new and interesting conversation.  I’m 100% confident I will lose my voice every day, but I’d like it to be from talking to fans about the gameplay, not repeating what the A button does over and over again.  I really have no idea whether anyone will read it at all.  I’ll report back on Friday.

I’m sorry I didn’t get this up yesterday, not only because it deprived you, the wonderful fans, of a day of plumbing its mysteries, but also because I literally finished it and hit send today, 11 minutes after the absolute latest deadline the printer gave me.  I didn’t even have time to read the whole thing.  I hope it’s not completely opaque. If I’d put it up yesterday I could have taken feedback and made it better, but as Donald Rumsfeld always said, “You go to PAX with the Four Page Instruction Manual you printed.”

Here is the print-ready PDF, which assumes you can print two-sided, and then you fold it in half to get the four-page booklet.

This document represents a fairly accurate beginner’s view of the state of SpyParty right now.  After you play a few times, you start to get to even more depth that I couldn’t fit on here (as if it’s not way too dense already!).   Let me know what you think in comments, or if parts are confusing, or if it makes sense.

Here’s a teaser of our booth while under construction this evening:

Monaco & SpyParty PAX Booth #3004, WIP

I’m having the two players face each other, and the big HDTV is duplicating the small HDTV closest to me taking the picture.  I’m counting on good sportsmanship from the other player by assuming he or she will not turn around!  And, hopefully the crowd won’t yell out spoilers.

It wants a little contrast adjustment and whatnot, but I figure the raw file is most useful.  Here’s the photoshop PSD:  xbox_360_controller.zip (24mb)

1. Yes, it’s basically insane to expect people to read documentation before playing a game demo at a convention, but the game is currently so inaccessible to noobs that I really have no choice.  I’ll upload the draft document here tomorrow so people can give feedback.

This is a screenshot from one of the insurance policies:

It’s good to know the relative value of these things…

In cable-trippingly related SpyParty news, I’m currently testing whether the warning on my 10′ USB extension cable, which reads, “Caution – do not exceed ten feet of total cable distance on low-speed devices, sixteen feet on high-speed devices”, is really accurate, or whether I can run an Xbox 360 wired controller (9′) over it.  It seems to work so far…famous last words.  Maybe I’ll pick up some USB repeater cables.

While searching, I found this excellent quote from the Xbox FAQ page:

9. Can I connect wireless and wired controllers at the same time?

You can connect up to four controllers in any combination you can think of with a maximum of up to two wired and two wireless controllers.

Any combination you can think of, except most of them.

Obligatory SpyParty screenshot:

"Would you care for a drink, General?"

"That's a knife..."

I’m not sure how well this claim would stand up in the face of actual data, but after James Cameron subjected humanity to the 3+ hours of Titanic and became King of the World¹, it seemed like the film people just gave up on parsimony and stopped leaving much of anything on the cutting room floor.  All the footage went into the movie.  No hard editorial decisions were made.  2½ and 3 hour films became fairly common, and nobody was making 90 minute films anymore.

Now, plenty of film buffs, critics, and people who have to urinate have debated the “length issue”, and I’m not actually interested in contributing to that here.  I am interested in pointing out that the debate rarely seems to center around the concept of “value” in terms of “money/time”.  It’s always about what’s the right length for the material², or did the director suffer from logorrhea³, or was the studio cynically trying to please all of the people all of the time, etc.  You don’t often hear people talking about movie prices tied to movie length.  Not that people don’t complain about movie ticket prices, mind you, it’s just that they don’t seem to couple them to the movie’s length very often.  People say, “movies are or aren’t worth $X”; they don’t usually say, “$X would be a good price for a N minute movie, but any less is a ripoff!”.

Other mature art forms also avoid this money/time value comparison.  People joke about how gigantic Infinite Jest is, but they don’t talk about it in terms of cents/page.  Should iTunes charge by the minute for songs instead of a flat $.99?  If so, Frank Zappa’s back-catalog would be quite pricey…

However, you see this “value debate” about game prices and game play length all the time.  In fact, it’s the usual way of talking about game value on the internet, as far as I can tell.

Why is this? What’s different about films, books,  and music, as compared to games?

If you’re familiar with my lectures and rants, you will see my answer coming a mile away:  I think it’s because these other forms deliver (or, at least, are clearly capable of delivering) deep and compelling emotional experiences, and it just seems gauche to break them down into money/time or money/size.  You can talk about the value of the painting, and everybody does, but you don’t break it down any farther than that—you can’t talk about the value of that flower versus the farmhouse, or the upper-left corner versus the lower-right—because you lose something ineffable in the analysis.

This topic came up most recently amongst a bunch of indie game developers after LIMBO came out on Xbox Live Arcade, and there was some discussion, and some more discussion, and we decided to do Size Doesn’t Matter Day.  But, it’s got a long history, especially with indie games.

The typical analogy made by defenders of game pricing and value is to the cost of eating out at a restaurant.  When the price being discussed is $15, the food being discussed is usually fast.

And, while it’s true you will pay more for a pizza these days than you will for a “AAA Indie Game”—or you will if your pizza is any good—and, yes, a $15 game will give you more direct hours of content than a $15 movie will, I claim if you’re even engaging at this level, you’ve already lost the argument.

So, while I think the focus on game length relative to game price is silly, I think the only way out is to make better, more meaningful games.  That is the most compelling argument we have against people who complain about $2/hr (Dragon Age or whatever @ $60/30 hours) versus $3.75/hr (LIMBO @ $15/4 hours).  Even when the economy is down, and you lost your job (hey, like me!), or you’re a kid trying to scrape together your allowance, or whatever, if we make games that strike deep emotional chords with people, that, and only that, will wash away the superficial discussions of value as defined by money/time.

Assuming we actually figure out how to do that, we’ll look back on this debate as an historical artifact, like discussing whether a nickel was too much to put into a Kinetoscope to watch the 5 seconds of Fred Ott’s Sneeze.

Links

Okay, here are all the Size Doesn’t Matter Day posts I know about.  Some of these are set to go live in the morning, so don’t report a bug until the sun rises.  Also, post a comment if you find more, and I’ll put them up here.

• Here is Jamie’s article on Gamasutra
• http://24caretgames.com/2010/08/17/does-game-length-matter/
• http://2dboy.com/2010/08/17/too-short/
• http://blog.wolfire.com/2010/08/Game-length-and-value
• http://www.brokenrul.es/blog/?p=314
• http://gamesfromwithin.com/size-matters
• http://macguffingames.com/2010/if-size-doesnt-matter-where-do-you-get-the-virtual-goods
• http://mile222.com/2010/08/a-haiku-about-game-length/
• http://nygamedev.blogspot.com/2010/08/coming-up-short.html
• http://positech.co.uk/cliffsblog/?p=810
• http://retroaffect.com/blog/160/Size_Doesn_t_Matter_Day/
• http://the-witness.net/news/?p=438
• http://www.copenhagengamecollective.org/2010/08/17/size-does-matter/ (potentially NSFW)
• http://www.firehosegames.com/2010/08/how-much-is-enough/
• http://www.hobbygamedev.com/spx/short-videogame-design/
• http://www.spyeart.com/misc/gamelength_MichaelTodd.html
• http://www.tunahq.com/2010/08/size-doesnt-matter-day/
• http://mindflock.com/2010/08/why-arent-there-more-middle-ground-games/
• http://www.enemyairship.com/2/The_Finite__Irreplaceable_Hours_of_Your_Life/
• http://www.lazy8studios.com/size_doesnt_matter
• http://www.next-gen.biz/blogs/the-long-game

Also, #gamelength on twitter.

Epilogue

Ironically, because SpyParty has a strong “online multiplayer competitive player-skill component”, this whole discussion is somewhat academic for me, for this game, at least.  My goal is to attain what I call “Counter-Strike levels of replayability”, which traditionally trends asymptotically towards $0/hr of “entertainment value”⁴.  However, my goal is for people playing SpyParty to want to do so because its engaging them in a deep and meaningful way, not because it’s a cheap way to spend time!

1. OMG I didn’t know (or blocked it out) until watching that again that Titanic won Best Editing?!
2. That link is particularly a propos since he disses video games offhandedly.
3. Yes, it’s a real word, and an awesome one at that!
4. Of course, the business end of our industry is going to try hard to fix that.

Well, now’s your chance! If you can make it to Seattle, WA on September 3rd, 4th, or 5th, 2010, and get to the Penny-Arcade Expo (otherwise known as PAX) and find booth 3004, you can play Monaco and SpyParty until the person behind you in the queue kicks you off the controller! No pushing in line!

Andy Schatz, the awesome indie game developer behind the awesome indie game Monaco (and winner of this year’s Independent Game Festival, that’s Andy there on the logo at that link!), and I have gone temporarily insane and gone in on a booth together at PAX, so gamers can come playtest Monaco and SpyParty themselves! Don’t take the press’s word for it, decide if you think the games are as awesome as they claim for yourself!

Monaco is a super fun stealth heist game with a lot of similarities to SpyParty, so they’re the perfect booth buddies! The two games are not only developed by solitary guys toiling away in their artist garrets as games should be made, but they also have very similar aesthetics and fictional themes, since old heist movies and spy and mystery movies are all close siblings stylistically, and both have awesome multiplayer.

We’re going to set up a nice comfortable space at PAX to play the games with your friends (or enemies), and we’re both rolling out a bunch of new features at PAX that nobody else will have played.

Jordan Devore at Destructoid wrote up a great piece about our decision to get the booth, and our preparations! Check it out!

Wow, Anthony Burch, ex-dtoid writer, now at Gearbox, just updated the HAWP blog with this super-kind post. Quote:

In honesty, I would not be surprised if SpyParty and Monaco were the best games being shown at the entire Penny Arcade Expo.

Yowza!

Here’s the map of the show floor where we are located, and you can see there are a bunch of other well-known indies in the immediate vicinity:

Monaco & SpyParty at PAX 2010

I’ll be posting more about what’s new for SpyParty for the PAX build of the game, as I’ve started with this post on the new “bookshelf mission”.

For now, please spread the word, tell your friends, become fans on Facebook (Monaco, SpyParty), and follow us on Twitter (Monaco, SpyParty)!

Indie games live and die based on word-of-mouth, so if you think the games are cool and want to support indie game development, do your part!

Thanks!

Truth be told, I’m just not worrying about the fiction right now; I’m completely focused on the core gameplay, and making it as deep and interesting as I possibly can, and I’m 100% confident that no matter where the design leads me, the results won’t be as absurd as Moonraker.

The old bookshelf mission mission was one of the more interesting missions in the SpyParty gameplay prototype, for a number of reasons. First, let me describe the mission, hopefully for the last time, since I just gutted it. :)

Here’s the current test level, called the ballroom:

Yes, it’s a box, with windows on two sides. In addition to ignoring the fiction, I’m also not worrying about level layouts very much yet! Every playtest you’ve read about in the press has been in this box. I’ve got the geometry for a couple more levels built, but haven’t needed to trot them out yet to keep things interesting, so they’re on the back burner for now.

Back to the bookshelves. You can see there are two of them in this level, one blue and one green. You can see some NPCs at the bookshelves, reading. I’m the woman in the plaid dress with the green triangle over her head. Don’t shoot me, I’m trying to explain something.

The NPC AI will occasionally decide to go to the bookshelves, take a book out, page through it for a bit, and then put it away, and go do something else. As the Spy, you can of course do the same thing.

The old bookshelf mission went like this:

• Your goal is to move a book from one of the bookshelves to the other. You can choose while playing whether you move it from the blue to the green, or vice versa, and that’s part of the strategy, since they have different levels of occlusion due to location and nearby people traffic.
• For this example, let’s say you start at the blue bookshelf.
• You can pick up a book like the AIs, and you’ll sit there and read it.
• While you’re reading it, you have two affordances², you can Put the book away like the AI, or you can Pretend to put the book away, but actually hide it in your jacket.
• If you put the book away, you look exactly like the AI when they put it away.
• But, if you hide the book, the animation is a little different. This is a hard tell, which I’ll talk about at length in another post, but it basically means it’s something the AI will never do, so if the sniper sees this clearly, you’re made.
• Once you’ve got the book hidden, you can go over to the green bookshelf. It’s not usually a good idea to beeline right from one bookshelf to another, because the NPCs rarely do that, so go talk to some people first, or look at some artwork.
• This is where things got a little overly complex. At the green bookshelf, you can either Pick up another book, in which case you have a hidden book from the blue bookcase and a book from the green bookcase in your hands, or you can Fake pick up a book but really take the hidden book out, in which case you’ll have a book from blue in your hands and nothing hidden.
• If you’re holding a book, then you have the option to put the book away.
• But, if you picked up a green book, you could then Swap the held and hidden books, so you’d have the book from blue in your hands, and a hidden book from green.
• Once the book from the blue bookcase is in your hands, whether it got there from a fake pickup and unhide, or swapping, if you put it away normally you’d accomplish the mission.

If you think that’s complicated to understand reading it here, you should have tried playing it. As I’ve said before, I’m wholeheartedly invested in the Blizzard Depth First, Accessibility Later development style, but man, you need people to be able to figure out how to put a damned book away while they’re playtesting your game. There is text displayed when an affordance is available, and one of them for this mission was “Fake Pick Up Book with Hidden and Hold It”. I can hardly parse that sitting here reading it, let alone when I’m worried somebody is going to shoot me in the head if I screw up.

It may have been a trainwreck from a learning standpoint, but the affordances were designed this way for a reason. The complexity was a direct result of the design goal that you always be able to “play it cool” if the sniper’s laser swept over you and he or she started paying attention to what you were doing at the bookcase. You always needed the option to do exactly what the NPCs would do at any moment (assuming you weren’t in the middle of playing a hide animation or whatever), and these affordances gave you that option, at the cost of clarity.

So, a redesign was in order, but…

Once people learned the state machine, it was actually a good mission with many interesting characteristics I wanted to preserve. For example, it was the only mission that took place across space in the level instead of at a point. It was two-part, with tells at either end, which made it hard for the Spy, but rewarding when you pulled it off. The Spy got to choose the order, so the Sniper was never sure which bookshelf to be watching for late in the match. It had a nice version of David Sirlin’s yomi layer concept, in that since it was hard, Snipers would assume Spies would not choose it. Of course, this made Spies choose it. It had fixed known locations where it would happen, so it could be camped, but while camping it, the Sniper was not looking elsewhere. Etc.

Yikes, it’s 4:23am. Okay, more on this redesign tomorrow…

1. Has anyone under 35 even heard of microfiche?
2. Affordance is a fancy design word for thing you can do.