I wanted to make a social game last year, so my girlfriend Brenda told her employers (Lolapps, Inc., a social game developer in San Francisco, California) that I was interested. I started consulting in August 2010, designed the game in a few days, and got the team working on what I decided to call Ravenwood Fair. At the beginning, I did art direction toward a painterly style until the lead artist, Christine MacTernan, totally got the look and took over from there. I created and coded all the sound effects, and worked with a great composer named Aaron Walz from Game Audio Alliance on the game’s looping soundtrack. He did an excellent job, as the song can be played infinitely without annoyance – the true test of a great composition.

The team was very small: four programmers, two concept artists, two animators, a UI artist, a producer, Brenda and me. That’s only 12 people. In addition, the game launched on October 19, which means it took 2.5 months to go from concept to ship. This causes many a social game publisher to gasp, but there’s a good reason for it: the team worked very hard, and they had already been working on several game concepts for 7 months before I showed up, so they were already “at speed” and no ramp-up time was needed.

When I arrived at Lolapps, the owners told me I could create whatever, as long as the game would make money and launch by September 30. That was less than 2 months by the time I started working on it. I knew I had to hurry up and come up with a concept, so I quickly identified the best game on Facebook at the time, FrontierVille, and played it for three days so I could understand its design patterns and monetization model (I had been following the social game space since 2009). Then, I got the scoop on the engine we were using and found out it was really fast and could draw a ton of graphics on the screen quickly. I learned from the lead coder how much code had been written for the previous game prototypes, so I needed to use as much as possible to keep moving forward and not go backward much.

The idea for the game appeared very quickly. The setting was going to be a fairytale/storybook/magical place that floated in the air and was occupied by a slightly scary forest. Since the biggest games on Facebook were business sims, I needed to have the player build their own little business, and I wanted it to be like nothing else on Facebook. I thought a Fair would be a really nice obvious positive goal for the player to build, and there wasn’t a carnival/circus/fair sim on Facebook……yet. My thought was this: when you’re building a place for visitors to come and occupy, why wouldn’t you want them to be doing something really fun while they’re there? What’s more fun than Disneyland? I imagined what it must be like at a high-level point-of-view to actually RUN DisneyWorld in Orlando, Florida – to have to carve out a bunch of wilderness packed with alligators and other denizens, and then have to build a fun place in the middle of that – and KEEP it happy and fun, all the while the alligators and beasts want their place back. I figured that would be a fun game to play – as I cut back a scary forest, I build a Fair, visitors come and use all the things I’m building, and I get to interact with them in a light fashion. That was the basis of my design, so we had our first meeting, and the producer and leads wrote down the long list of things that we needed to start working on immediately.

After August, we were still working on the foundations of the game systems and entire art overhaul, when Brenda joined the team from Critter Island (it had been shut down by Facebook). Brenda immediately started working on the quests, the map layout, and the tutorial parts of the game. She had to basically try to get all the final parts of the game together while I was still working on the foundational stuff, and we had to make sure that they would meet in the middle and the game would be complete by September 30.

Near the end of September, to make the launch date, the game had been re-scoped and re-prioritized so much that it wasn’t going to be anything I’d want to play if we shipped it in that state. Facebook games are well-known to ship incomplete and get fixed post-launch. I’m not from that world – I’m from the world of “You only launch once” and the first impression is extremely important. There were game systems missing that absolutely needed to be included in the game for the design to harmonize, and it got very tense at the end of September with the execs pressing for launch and me holding the game back to make it great. Every day in October was a “Can it ship?” day. It took 19 more 12pm-to-4am days to get the game in a state that I’d launch. In fact, I held back on doing interviews about the game until I knew it was going to be something I’d put my name on. Finally, around October 15, the game was quality enough for me to do the Dean Takahashi interview in VentureBeat.

The game launched, but the team kept working. We had lots of holes to fill, and obscure bugs to fix. The launch on Facebook went as good as a launch could possibly go: millions of people played the game and hundreds of thousands still join every week. Brenda and I ended our consulting with Lolapps on February 25, 2011 so we could concentrate full-time on our new startup, Loot Drop, Inc.

Just recently at the beginning of March, Ravenwood Fair made the #15 slot on the AppData Top 15 Applications list – this list is of ALL applications on Facebook, not just games (over 125,000 apps are on Facebook). Ravenwood Fair is the #1 non-Zynga game, and that’s quite an accomplishment. Currently, over 25 million people play Ravenwood Fair every month on all social platforms (over 11 million on Facebook alone), and over 1 million play every day on Facebook only. I never thought I’d see the day that I would make a game that doubled World of Warcraft, player count-wise. I’d like to thank Brenda and Lolapps for giving me the opportunity, and the Ravenwood Fair launch team that worked so hard on the game.

Ravenwood Fair Trivia Time!

                          

After three months, Tom Hall officially joined us in May 1991. He wanted his transition from Softdisk to go smoothly, so he stayed to train his replacement. Our first year at id was spent mostly developing games for Softdisk’s Gamer’s Edge subscription disk – the product I started, but then left when the company could hold us no longer.

In our first year, 1991, we created Shadow Knights, Dangerous Dave in the Haunted Mansion, Rescue Rover, Hovertank One, Rescue Rover II, Keen Dreams, Commander Keen: Secret of the Oracle, Commander Keen: The Armageddon Machine, Commander Keen: Aliens Ate My Baby Sitter!, Catacomb 3D and Tiles of the Dragon. Eleven games in one year! Not counting the Keen games, Softdisk paid us $5,000 for each game. It was a good thing our first Commander Keen trilogy was making money and gaining each month, otherwise our trajectory would have been drastically different. (In the halls of id Software, some of these games are on display in hanging frames.)

When we started creating Wolfenstein 3D in January 1992 (in Madison, Wisconsin), we still had one more game to create for Softdisk, but we didn’t want to do it. We were too excited to create Wolfenstein 3D! So, George Broussard, co-founder of Apogee Software/3D Realms, offered to create the final game for us. The result of his labors, using our game engine, was the rare and obscure game, ScubaVenture: The Search For Pirate’s Treasure! (An interesting trivia point about the game’s title: one of my programming heroes, Nasir Gebelli, wrote a game called ScubaVenture back in 1983.)

Our first year was the last time we were so prolific. As the games got more complex, and better, the time to develop them grew. Wolfenstein 3D took 6 months, DOOM took one year, Quake took 18 months. In between, our sequels took less time (Spear of Destiny, 2 months; DOOM II, 8 months), but they were using existing code we’d already written.

Twenty years, wow. That is quite an impressive accomplishment, due in large part to the torch-carrying efforts of John Carmack and Kevin Cloud, both of whom remain at id Software today. May id Software live another strong 20 years!

                          

Well, I think it’s awesome for several reasons. First, because I’ve been a huge Duke fan since Duke Nukem 3D. I absolutely loved that game. My memories of playing DN3D are as awesome as my memories playing Dark Forces, Outlaws, Jedi Knight, and Ghost Recon. Secondly, I’ve been friends for years with Randy Pitchford, the man who bought the Duke IP from 3D Realms; to carry the torch as it were. I’m really proud of Randy for doing that – Duke is a great character that deserves another major release, and I’m interested to see what directions Gearbox will take Duke in the future.

I visited 3D Realms in May 2000, right after shipping Daikatana. I went over to check out the latest on DNF. George Broussard showed me the game, and it was seriously awesome. Granted, I was seeing the best stuff they had available at the time, but what I saw was really great. I was excited to play the game. Alas, development of the game went through many changes, and not having a dedicated producer early on took its toll. When I heard 3D Realms had halted development of the game, I was half surprised, and half not. It had been 12 years, and that’s an incredibly long time to be in development – I mean, it’s really incomprehensible. I can’t fathom what it must feel like working on a game that long, then not releasing it, especially if you were there at the start of development.

But now we can all breathe a sigh of relief, those of us who kept the excitement at bay for so long. The Duke Nukem Forever development team that continued creating the game, even after being laid off, without being paid, for the pure love of the game, deserve an epic amount of applause and attention. These guys are the heroes. And I’m still excited to play DNF.

                          

There are some other great articles inside, such as Ensemble Studios’ development transition from Age of Empires 1 to 2 (Age of Kings), a short interview with my friend Randy Pitchford, and an article with ideas about how to you can spend your time if your game has been canceled.

The people who put this magazine together are a great bunch. They have many, many years of experience in the industry and want the magazine to be only the highest quality. I worked with Jessica Tams and Jake Simpson, and they are both very kind and generous to include me here. Gamesauce is only for professional game developers and those in the game industry.

You can read the entire magazine right now.

                          

I have several pages of notes that describe a game different from the Quake that you know. I’ll put these up when I start the Quake game page so I can explain what I was thinking for that design.

I’ve also put up lots of other pictures in many other sections. The smugmug site is really great – you should try it out for your own collections. You can totally customize it however you like.

                          

Next year will be two decades of Keen, Wolfenstein, Doom, and Quake. How many game companies still standing can boast 20 years? Not many. And most companies that live past 20 years are so far removed from their origins that they’re not even the same company.

John Carmack and Kevin Cloud, two of id’s earliest team members, and John as a co-founder, are still with the company, and working hard on RAGE and Doom 4. John is right down in the pit with the development team, where all the action happens. Kevin manages, as always, exceptionally well on multiple fronts. I salute their efforts to continue the dynasty.

It all started on February 1, 1991. John, Adrian and I left our jobs at Softdisk (R.I.P.) and began work immediately on Dangerous Dave in the Haunted Mansion at the id lake house in Shreveport, Louisiana. Tom Hall needed to stay a few months longer at Softdisk, to help them ease his transition out of the Apple II team. But after work, he’d come over and help out.

Those were prolific years. The years 1990 and 1991 saw us develop almost 20 games with wildly different play styles and themes. The Keens, The Dangerous Daves, the Rescue Rovers, the Catacombs, and more. There were a lot of them and all were fun. Wolfenstein 3D begat the first-person shooter genre in 1992, and id hasn’t looked back.

Congratulations on 19 years, id!

                          

DOOM development on NeXTSTEP

Here’s as much information as I can glean from this scan:

• it’s John Carmack’s NeXTStation; we had a NeXTCUBE which was used to scan the videocam images we made of our monsters (clay and latex)
• the cookie monster icon is for John’s app, Fuzzy Pumper Palette Shop, which we ran on the Cube and processed the images that the videocam captured. The videocam was plugged directly into the Cube.
• the screen is mainly showing the debugging environment of SuperDebugger as DOOM is being debugged
• the DOOM app window has a black, draggable title bar with the 320×200 game screen rendered within it
• the Objective-C compiler is just under the game window
• at the bottom of the screen are icons representing the running apps; many are source files

About the article, here are my clarifications and bug fixes:

• the reason Scott Miller was trying to get in touch with me was because of my PC game (port) Pyramids of Egypt, a game i wrote originally in 1985 on the Apple II and ported to the PC in 1989. No one else was involved in making this game.
• John, Adrian and myself did NOT quit Softdisk after we started making Commander Keen; we did it all after work and on the weekends at the lakehouse
• Tom Hall was included in this group and it was mainly John C., Tom and myself that made Commander Keen 1-3. Adrian did a few graphics for Keen 3 at the very end of the project but he wasn’t involved at the beginning
• One month after releasing Keen 1-3 we received our first royalty check and decided to quit Softdisk and form id Software on February 1, 1991
• id was located in Madison, Wisconsin between September 1991 and March 1992; this was when i started getting involved with Raven Software as they were local to the area
• id moved to Mesquite, Texas in March 1992; at this time, Jay Wilbur and Kevin Cloud joined the original team: John Carmack, Tom Hall, Adrian Carmack and myself
• DOOM was never originally going to be “Green and Pissed”; that was a game idea Tom had before we made Wolfenstein 3D and i thought it was a hackneyed idea (science lab experiment goes bad and mutants run amok) so i came up with the idea of a Castle Wolfenstein remake
• when Tom left id he immediately started working on Wolfenstein II – we licensed Apogee the Wolfenstein name to make the sequel but later decided not to continue with it…..so Wolf II became Rise of the Triad
• id’s taking over distribution for DOOM is described as a “staff split” but Apogee and id were always two seperate companies
• the “medium detail” mode mentioned in the article was called “high-color mode” and was only available in the first version or so of DOOM and was removed because of its incompatibility with some video adapters
• bullets aren’t physical projectiles in the DOOM engine
• we only put in multiplayer near the very end of the game even though we announced the feature in our January 93 press release about DOOM
• the graphics for Wolfenstein were mostly drawn by Adrian Carmack but Kevin Cloud helped a lot after he came onboard on April 1, 1992
• when we did video capture of our latex and clay models, the video camera was pointed at the model and a live video feed was plugged into the NeXTCUBE which, at any time, we could click a button to take a frame grab. This frame was then fed into our program Fuzzy Pumper Palette Shop which took it from 24-bit color down to a 256-color VGA paletted graphic. At that point it was saved on our network where Adrian and Kevin could load it into Deluxepaint II and clean it up.
• during Wolf3D’s development we came up with the idea of a contest to find the hardest thing in the game; for beta, we drew a graphic that was NOT going to be the final code, so just before mastering we changed the graphic – we had never been “plagued by leaks”; an alpha version of DOOM *did* leak however
• Wolf3D was banned because of Nazi content in addition to violence; in Germany there is a sheet of banned imagery that will put you in jail if you are seen with it on – Wolf3D was full of this imagery (Nazi imagery)
• Cygnus Studios became Rogue Entertainment after an internal revolt; they were working on Strife for us
• a SNES version of DOOM *was* created in secrecy by Sculptured Software; they presented a final, finished SNES version of DOOM and sent it to us asking if we wanted to publish it; oh hell yeah!
• the Jaguar version of DOOM was done internally at id; it was very fun and challenging but the Jaguar died a slow death

Great article!

                          

The error I got when I tried running the code was this:

1180: Call to a possibly
undefined method xxxx

This basically means that Flash couldn’t find the class that I created. But I didn’t get any errors from the “import” statement that references the .AS file that has the class in it (inside a package{}).

If I changed the filename I’m trying to import I get an error. Ok, so it knows the file is there but it cannot find the class inside the file for some reason.

I searched all over the net for an hour trying to find the answer to this problem. NOWHERE DID I FIND THE SOLUTION THAT I AM ABOUT TO GIVE YOU. The solution to the problem is something that, evidently, Flash programmers JUST KNOW from experience but the solution isn’t written anywhere that I could find on the internet and my Flash coders don’t know how they know the technicality I’m about to explain.

So it comes down to this: the way you write the name of your class MUST be the EXACT way you name the .AS file it’s in.

For example: if I create a class named BugJar like so:

package
{
  public class BugJar
  {
    //...code...
  }
}

then I **MUST** save it in a file named BugJar.as!!!!

Inside my framecode I would have:

import BugJar;

…and it would work.

This is just completely disgusting to me as a programmer. Linking the name of a class to the filename of the file it’s in is just wrong.

What’s worse is that Flash doesn’t even give you the proper error for this:

filename: bugjar.as

package
{
  public class BugJar
  {
      //...code...
  }
}

framecode:

import bugjar;
var jarugJar = new BugJar();

The above example WILL NOT WORK and Flash doesn’t give you an error message that helps you understand or fix this problem that shouldn’t exist in the first place.

                          

What else has been going on? A ton, especially at work. We’re over 60 people now and working away on our MMORPG. It’s a lot of fun but also an insanely interconnected web of dependencies and just a huge project size, the biggest I’ve done thus far. We’re still in stealth mode for quite a while yet so I can’t say more yet but we’re all very excited by our progress!

My iPhone continues to be the paragon of electronic goodness. It just rules, completely. I love watching TV shows and movies on it, taking photos, listening to music, getting email, sending SMSs, etc, etc. It’s AWESOME! Rhaluka loves hers as well.

I’m slowly working on Quinn Fox’s official site as well. I gotta get his music up there.

Other interesting things going on: a Daikatana recreation in the Half-Life 2 Source engine; I might be the star of a reality show (these things never work out though; I’ll be in China for a week in November; Dangerous Dave on cell might be published soon.

                          

Right now, in the news section on Gamasutra is a blurb about the interview (with link). And it’s so great to see my story just under GAMECOCK’s story…..LOL. So perfect.

Anyway, here’s the interview.