A I   C a f e
Cafe logo

 

 

 

 

 

 

 

 

 

 

Cafe art

 

 

 

 

Home | Tutorials | Tutor Bot | Message Board |
Links | Editorials | Archives | SkyBlazer |
Minion | Minion Arena

 

Size Does Matter Jan 26

      It'll be done in two weeks.
      Everything will be new.
      Mine is bigger than yours.
      These are all statements made to describe a mod. They have all become Quake cliches. And none of them have any meaning.
      Many developers want to boast about the size or scope of their mods. Yes, we love to see total conversions. Yes, size does matter in modmaking. But size here means not how large a project is, but how small it is.
      I refer specifically to a relatively new entity in the user community: the mini-mod.
      Prime examples of these compact projects can be seen in Akuma's Hurried Descent, ze0's Fist Fight, ELF's many patches, and Void's FrikBot modifications. Heck, even the Tutor Bot -- with its whopping one file -- is a mini-mod.
      All of these mods have one thing in common, one thing that makes them very cool indeed: they're done. The guys at ELF, Skinski and harbinger of f!sh, were the first to turn me on to the idea. They taught me (and anybody who visits their download page) that a small mod is a finished mod.
      I think it was Skinski who said he never starts a mod that he can't complete that weekend. Sounds like a simplistic statement, but it may be infinitely wise. After all, you, me, and everyone in the Quake community has more ideas than time.
      Many of you already have compact mod ideas, so this is nothing new to you. But there exists another facet of the mini-mod that makes it distinctly enjoyable. Simply put, it's a map.
      You know and I know that there is a world of difference between some little patch that adds a new weapon and one which possesses its own level. You can walk around in a level. It is a physical place in time. It is the body of a mod. Minion Arena came with its own levels, and it was a smash hit. SkyBlazer did not possess new levels, and it was overlooked by many gamers.
      Moreover, could you imagine MA without those maps? When you were looking at MA, you weren't seeing its rules, or its gameplay; you were seeing its maps. In a matter of speaking, you were inside Minion Arena.
      So, my advice to you: learn how to make a level. I don't care if you're a programmer, an idea man, a modeller, an AI guru, a mod producer, whatever. Get Qoole or WorldCraft. Learn mapmaking. Hey, I hate it, and I consider myself bad at it. But my best projects -- released and unreleased -- would not be the same without levels. The mods we remember the most are the ones we can see.
      We now have made two points: keep your mod small, and give it a map (or two). This gives us a pretty clear picture of the mini-mod. In fact, it gives me a rather specific image of one; when I think of mini-mods, I think of early video games.
      Think about it. The first arcade games and Atari titles featured one screen, simple rules, distinct gameplay, obvious originality. And fun. They always served up a large heaping of fun. Sounds like a good recipe for a Quake mod.
      Imagine the following titles set in the Quake world: Pacman, Donkey Kong, Pong, Berserk, Q-Bert, Gauntlet, Atari Football, Pengo. And you know, even these early Commodore 64 and Nintendo games might make cool mods: BallBlazer, Archon, Spy Vs Spy, Bomberman.
      Of course the concepts would have to be altered to fit the realm of Quake, but I think each of these is a great starting ground for a mod. Back then programmers were forced to think in simple terms, work with limited resources, focus on fun gameplay, and get done in a short timeframe. Titles revolved around running, jumping, and shooting. One programmer could create a whole game by himself.
      Sounds really familiar.

 

      Coincidentally, after I finished this editorial I discovered that Qoole 99 is now freeware. This is the full version; no brush limits, no inactive features. In case you didn't know, Qoole is the faster, easiest level editor out there, one which even includes tutorials on its site. I downloaded it. You should too.

 

An Exciting Time Nov 28

      Some people may think Quake One is old and outmoded. Some guys might think Quake development has dried up. Perhaps they think Quake is done.
      Well, they're f#@&king wrong.
      Just like Mark Twain said, reports of Quake's demise are premature. Classic Quake cannot be stopped, and it cannot be killed. There is much heated debate about this, as seen in the "I know why Quake is dead" and "What will happen after Quake 3?" threads on the Cafe message board. There are plenty of interesting opinions there.
      You probably remember when Quake 2 came out: most Q1 projects were put on hold as the q-mmunity checked out id's new creation. Most predicted that Q1 would die right there and then.
      True, many large news sites turned their attention to Quake 2. But like the media in Real Life, the Quake news pages cover some topics, and ignore others. Just because Q1 news is hard to find does not mean it is not happening.
      Many are now saying that the release of Quake 3: Arena will kill Q1 and Q2. All we have to do is compare Q2 development (relatively small and over-rated) to Q1 development (still benefiting from QuakeC) to know they are wrong.
      Obviously Q3A will affect the Q1 community, no doubt. But this mainstream game will also bring in new fans who have never been exposed to Q1 and Q2, never seen all these sites, never played all these mods. And I bet you that they will be more than a bit curious about that easy little scripting language called QuakeC.
      On top of that, there is still much Quake One development going on, whether you know it or believe it or not. The deathmatch maps that have been released in the last year or so easily beat those of the early days. Developers are applying new mods to the FrikBot, the Reaper Bot and the FrogBot. Someone has created a Q2-to-Q1 conversion. And there are even brand new total conversions like Defeat In Detail and The Nehahra Project.
      This is why I continue to update my links page; you may have to dig around a little more nowadays, but Q1 is still alive and kicking ass.

 

Wacky Mod Ideas Nov 7

      If you are involved in the Quake community, chances are that you have a few ideas for your own mod. If you are like me, then you probably have a few hundred ideas for mods.
      Since I started messing around with this stuff, I have come up with tons of possible mods. In fact, I have a whole notebook filled with dozens of proposed mods, each with detailed outlines. Not surprisingly, these are all single-player, bot-oriented games (id software would kill to get a hold of that notebook).
      Moreover, these ideas are actually pretty good; it's just that I merely do not have the time to code and craft them all. Everyone has more concepts than time. I'm sure you possess enough of your own ideas, but following are some of my incomplete mod proposals:

      Bodyguard. Several tough assassin bots are spawned, as well as one slow, axe-wielding "president" bot. It is your job to escort him around the level and protect him from the mercenaries. In alternate modes, you may be able to play as the president or an assassin. You might have relevant equipment like a sniper rifle or motion sensors.

      Release. I may have referred to this concept before. It's based on the childhood running game. One base, two teams. The "in" team tries to capture the "out" team and lock them in the base. The free members of the out team can run in and touch the base to "release" their imprisoned comrades. A little like Jailbreak.

      Pool. A crazy idea for sure, the map for this mod is actually the top of a huge pool table. Naturally, it features nine huge pockets and nine huge, um, balls. Like QPong for Quake 2, you try to push or shoot one of these balls into a certain hole. The maps would actually be easy to create. Wacky, yes, but lots of potential for humor and excitement in this one.

      Dodgeball. Each player begins with only an axe. A big ball (recurring theme here?) is created somewhere in the level. Your objective is to get that ball and hit the other guy with it, just like in gym class. The excitement would come from all the running, chasing and dodging.

      These conceptions are not revolutionary or anything. On the other hand, they would be relatively easy to do. I know you likely have your own ideas, but if you wish to create a mod out of any of these, be my guest. Just drop me some email to let me know; I would really like to play it.

 

The Best Damn Bot Mod in the World August 5

      When you sit down to create that bot game you always wanted to write, you must acknowledge the following. There are only two kinds of projects you can do: the best damn bot mod in the world, or one you can finish.
      That is not just my normal facetiousness, it's the truth. When I speak with guys on ICQ, I hear quite a few bot ideas. Intelligent monster communities. Artificial breeding. Bot that eat, sleep, and drink.
      The guys who tell me these things mean well; they want to do something good, and they have thought it through. They normally describe their visions in copious detail. They talk to me for hours.
      And then I never hear about the project again.
      Sometimes, I never even hear from the guy again. They desired to make the first type of bot mod, the best in the world. But their imaginings were so complicated, so ambitious, that they could not see them through. They forgot that rule which applies to most things in life, including AI projects: Keep It Simple, Stupid.
      I will tell you right now, the best bot mod in the universe was never finished. If you are a budding mod designer, you must set realistic goals and stick to them. Choose a game which you can finish. Release it. Get famous. Improve it later.

      I too have fallen prey to overly ambitious mod ideas. Minion was originally going to be a long series of games, with monster clans, epic themes, and radical gameplay. I have numerous Quake subdirectories featuring playable yet unfinished projects. And my notebook is packed with dozens of fully outlined bot mods.
      However, I have in recent days dropped the epic mod fixation. I was talking to Harb!ngeR of Fish recently, and he told me of a vast game he was developing. Yet his brother, SkinSki, said he refused to work on it. I asked him why. He said something like he wouldn't commit to a project that he couldn't finish over the weekend.
      I thought about this. I thought of all those big, incomplete mods sitting on my hard drive. I concluded that he made the right decision.
      You know, I built Minion Arena, a mod which people still download two years after it was released, in one week flat. The maps, the AI, the dialog, everything was fully completed in seven days. Minion took about six months to write, to develop, to code and recode, and it's not half as cool as Minion Arena.
      Maybe you should throw your 70-page design document away. Small AI mods, like Demon Tag Bots, can be quite fun and easy to build. In fact, there are several specific guidelines that can help us write a project today. Let's look at those now.

      1. Pick something the bot is good at. We know navigation can be hard for computer-controlled opponents, so avoid mods like Level Racing. A bot is proficient at simple tasks, calculation, and fighting.
      2. Alter the main elements of the game engine. These are maps, graphics and opponents. Changing one of these is good; converting all three is superb. Designing new levels has to be the most concrete way of creating a new experience. Offering new textures, models, or weapons can also be rewarding to the player. And improving the opponent intelligence might be the single most important aspect of the mod.
      3. Create new goals. The only objective to deathmatch now is to kill the other contestants. Leave this behind. Think of new aims, like kicking a ball into a goal. Imprisoning your foes. Capturing a special item. Guarding a particular entity.
      4. Adopt a specific theme. There is plenty of inspiration on the net already. Look for maps, player skins, or weapons that would make good games. For instance, you could make the Fragtown city levels into a mobster mod. You could take the Predator skin and the Alien model and build them into an Aliens vs Predator project. Or you could grab the BFG, plasma rifle, and similar guns and create a Doom Bot with them.
      5. Change the fundamentals of deathmatch. These consist of running for the biggest gun, collecting ammunition, and shooting. Remove the guns. Force the player to buy his ammo. Add more enviromental interactivity. Do something, anything to make your mod feel fresh.

      Examine these suggestions to form an idea for your mod. Then stick to them as you are doing your work. You will end up with a lean project that is easy to complete and fun to play.
      After all, we can't play ideas. And we won't have any fun with your mod until it moves from c:\quake\work to ftp.cdrom.com\public\quake.

 

The Action
Contruction Set
July 2

      I'm disappointed with the gaming companies.
      Oh sure, they are finally making progress in first-person shooters after years of stagnation. True, games like Half-Life and Unreal Tournament will end up as two of the best games ever made. But I'm still disgruntled.
      You see, I believe that QuakeC was as important and as ground-breaking as those titles. After all, that simple scripting launguage shaped the Quake community, and the Quake community shaped the entire online gaming world. QuakeC was, arguably so, a revolution.
      But you have to wonder where that revolution went. The Quake scripting language was dropped after only one game. And you don't have to be an MIT student to see that the murky, messy Quake 2 .DLL source code strongly resembles the Florida Everglades.
      I thought that game companies would look at the smash success of QuakeC and make it even easier and more enjoyable to edit games. Yes, you have UnrealScript, which is reportedly cool. But it is one dim light in a big, dark room.
      Call me naive, but after Quake I envisioned an "Action Game Construction Set." I am of the era of the great Commodore 64, for which I owned Adventure Construction Set, Pinball Construction Set, even Racing Destruction Set. These are all now legendary titles which shaped the industry. And they allowed me to create my own worlds.
      First-person shooters are perfect for editing, and the "AGCS" is a logical idea. A company would create a decent 3D engine, a good map editor, and a model maker as well. There wouldn't even be a scripting language, because you could edit the behavior of the entities in either program. In addition, they would throw in dozens of free maps, models, and sounds.
      Alas, the game companies don't share the same vision. Maybe this software would take too much work. Maybe there wouldn't be enough profit in it. Maybe they assume that the gaming audience is too dumb to edit stuff.
      Some may argue that we are moving forward, that DLL coding allows us to edit games like never before. But they misunderstand QuakeC. What if id software would have released the real source code, the raw code, to Quake when that game came out?
      What do you think would have happened? You have to doubt if we would have gotten the Killer Quake Pak, or Slide, or Minion Arena. I know for a fact that we would not have ever seen the last one, because it was written by an amateur, non-C coder, like most other mods on cdrom.com.
      In other words, only obscure, hacker-type guys would have edited Quake. That would've been cool, but it would not have inspired a revolution.
      No, QuakeC was not popular because it allowed us to edit the game. It was popular because it was so easy, so accessible, so fun. And if .DLL coding is so powerful, why can't I find one single player Quake 2 mod?
      It's a shame the revolution ended the day it began. It's a shame I can't go to the store and buy the Action Game Construction Set.

      In addition to this heartfelt editorial, I have a bonus for you. It's a very decent tutorial on creating bot classes. Develop bot types like warriors, ninjas, scouts, whatever you want. Two thumbs up for this one.

 

Where Are The Algorithms? May 28

      I can just bet you that some people who have visited this site have said, "This is the AI Cafe, so where are all the algorithms?"
      For quite some time now, I have followed several bot sites, message boards, and forums. I have learned something from all this research: most bot programmers are fixated on navigation.
      Of course, navigation is an extremely important part of any deathmatch bot. Learn as much about it as you can. But please, if you do me just one favor ever, just don't obsess over navigation.
      This may be heresy to some bot developers out there. After all, many of those sites that I follow possess an air of the technical, as if they want to mystify artificial intelligence, as if they want to keep it for computer science majors only.
      (A monkey at a keyboard could have eventually come up with the AI in Doom.)
      Navigation algorithms are great and good and all. But with the limitations on entities and speed in QuakeC, they usually don't help us much.
      Besides, you can have your Floyd's algorithm, you can have your A*, you can have your robust heuristics. But all of them don't affect the bottom line of map navigation: the bot has to get out of the room first.
      That may sound as if I'm trying to be facetious or sarcastic, but I'm not. Honesty. Put yourself in observer mode and watch any bot, the Reaper, the Omicron, the FrogBot, and you will see what I mean. A bot's main limitation in map exploration is finding his way out of the room.
      Let's look at it another way. A bot always looks for items, and of course, he oftens finds some. Right there, we can say that if a bot can see a thing, he can usually get to it. Even if he can't, we can give him a lot of simple subroutines to help him. The thing to remember is that he can see that item, which means it's likely close by, which means its likely in the same room.
      But what after he gets that item, and doesn't see any others? This is when even the best bots start to wander. And unless you have planted the entire level with waypoints, shortest-path algorithms and such won't mean squat for your bot.
      Another reason why algorithms are lame is because deathmatch is lame. That is, they are so rooted in the old rule of deathmatch, which is: run around and look for a big gun, and shoot someone with it. So a bot who finds the rocket launcher first is the smartest bot.
      Bo-ring.
      It's okay if you have your own opinion; you're wrong. The gaming world is leaving Doom-style deathmatch behind. Capture the Flag, Team Fortress, Domination, King of the Hill, Search and Recover. These are the styles of gameplay that we --and bots -- are beginning to play.
      Don't get me wrong, bots who can find shortest paths are wonderful. If you can do one, go for it. But their authors have usually spent months on accomplishing this one, simple, technical fact. Tell me, do you think you could tell the difference between a bot who always finds the rocket launcher and one who always starts with one? You must remember another main rule of artificial intelligence: the best AI cannot even be seen.
      Moreover, those smart bots don't usually do anything else. I mean, they usually don't have a lot of extra features, or new weapons, or new rules. They are just deathmatch bots. I played deathmatch bots in, like, 1997; I want something fresh by now.
      Look at these examples: Killer Quake Pak, Requiem, Paroxysm. The bots in those mods are not as intelligent as some of the others out there. But they surely are more fun than those other bots.
      At this point, you may think I'd rather give my bot a weapon, which, admittedly, I did in SkyBlazer, rather than teach him how to get it. That's not true. But my time is short and my patience is thin; I would rather spend more time on more exciting AI, especially when no one else is. For instance, in SkyBlazer, I taught the bot to find a ball and carry it into a goal.
      That brings us back to the bot being able to find his way out of a room. This is why I created my coffee_move() subroutine. Simply put, this gets a bot out of a room, period. And it's about five lines of code.
      The best thing your bot can be is mobile. Yes, mobile. As long as he can keep moving around, keep entering and leaving rooms, he will find a diversity of items, and the player will see him in a diversity of locations.
      It takes just several hours of accomplish this. But it would take you several months to improve on the navigation we've seen in the current smart bots.
      It is natural for beginning bot programmers to work on navigation, since it seems like the first thing to work on. But when I see whole threads on message boards fixated on this, I just think: too bad, these guys are just gonna work on this for months, get frustrated, maybe release what little they have, and quit, and we'll never see anything interesting from that.
      I'd rather have you start with some existing bot source code and teach it to play Team Fortress or something. So go ahead, work on your navigation. Just prioritize your time on navigation and things that the player will actually see. Also, remember that this isn't a Carnegie Mellon graduate project. Do something fun.

 

The Dilemma of Modmaking

     The best Quake mod was never released.
     As a matter of fact, the best Quake mod was never finished. No, in fact the best Quake modmaking team was never formed. And the members never even met each other.
     Ever since I started this site at PlanetQuake, I've been on the ICQ chat network (at 2716002). And ever since I got on that network, I have been amazed at one thing: nearly everyone who has played Quake has a good idea for a mod. Most of those guys began work on their projects.
     But then most of those guys never completed or released their work.
     That's a shame. A real shame. All of us single players have lost out on plenty of good mods for the simple reason that they're pretty damn hard to finish.
     There are good reasons for that, of course, and it's nobody's fault. It takes hard work to fulfill a project. One guy can't do mapping, modelling, marketing, and coding while remaining sane (I should know). There are at the very least four full-time jobs right there.
     On top of that, some of us are only good at one of those things. For instance, my AI coding is alright, but my maps are as much fun as a block of wood.
     Which brings us to the dilemma of modmaking. Even if you are talented at one of these game aspects, and even if you do a project based on that, it's just a patch.
     Yes, I've always made a distinction between "mods" and "patches." A mod has to have thre main aspects: new locations, new asthetics, and new objectives. In other words, new maps, new graphics, and new goals.
     A patch is just one of those things. There are a million patches out there. A map, a gun. That's just an accessory. It's not a new game.
     This problem has handicapped me from the start. I have a notebook full of ideas and a hard drive full of bots, but I could never create maps. Take SkyBlazer for instance: a perfectly good idea but nowhere to play it. In fact, Minion Arena was a small, dumpy game compared to some of the plans I had.
     Nowhere is the patch problem more evident than on cdrom.com. That's a patch jungle. Say you download a new chaingun model. You load it up. You shoot it. Okay, then what? You got nothing to do with it, except maybe kill two-year-old Quake monsters.
     But of course there are some damn cool patches out there. Remember, I said it was a dilemma. I've seen some maps and models at cdrom.com that were better than anything from original Quake.
     (So many times I've visited there and yelled at myself, remarks like "Damn, I wish that map designer would have made maps for my team bots!" or "Shit, I should've done bots for that deathmatch game!")
     Put another way, if you could scoop up all the patches there and organize them into one big game, it just might be the coolest mod ever.
     Yeah, that might be ambitious. But I'm really talking about all you guys out there working on your different talents, perhaps not finishing your games. And I know there are even more of you hesitant to start a project because you're not sure how it would be finished.
     Well, I'll say to all of you that you wouldn't believe how many other guys are still trying to do what you're doing. Still trying to wring life out of Quake two years after it was released.
     If more of you mod developers joined to together, you could complement each other, finish your projects more easily, and perhaps produce the next great Quake mod.

 

Star Wars: The Phantom Mess May 23

      I love movies.
      Old movies, new movies, drama and comedy. I love movies as celebrated as Spartacus and as cheesy as Creepshow. I watch films from China to Sweden. I read trade magazines like Empire, Cinescape, and Neon. I think I know a little about movies.
      The Phantom Menace sucked.
      I won't say what all those would-be web critics say, that "the characterization was bad." Everyone says that about every genre movie. In fact, I probably won't say anything original.
      But I will say this: the plot was thin and confusing. Lucas cuts his shots way too quickly. The film played like an annoying video game. Darth Maul was completely wasted. Half the characters had accents I couldn't understand. The cast had no chemistry among themselves. And the Gungans were, yes, even more annoying than Ewoks.
      I was there in 1977 when Star Wars came out. I don't think many people who read this site were. But I'm sure you've heard my generation ended up hating Ewoks, hating the cutsiness, longing for the dark drama of The Empire Strikes Back. I guess most younger people don't feel that way, and I know Lucas doesn't, because the Phantom Menace was one of the most irritatingly cutesy films I've seen.
      What's more, I don't see how anybody who enjoys films like Clerks, Pulp Fiction, and The Killer could enjoy the new Star Wars. It was not even a kid's movie; it was a kiddies movie. It didn't even set up any interesting plot points for the second one.
      Sure, you'll hear many different opinions. But I think that a few people will say they liked it just because they wanted to like it, as I did. I was even surprised at my own reaction to the movie.
      That's too bad. As a Star Wars child, I'm in mourning.
      Anyway, I just had to throw in my two cents, since I paid my seven bucks. I do have a good tutorial for you. This one reintroduces my powerful and versatile bot velocity system. One of the most highly recommended lessons yet.

 

Coffee presents
The Top 10 Most Wanted
March 19

      I'm an avid Playstation gamer, and I read a ton about games. Every month, the Official Playstation Magazine features the "top ten most wanted games," which of course are the hottest upcoming titles.
      That's a really cool idea which I thought could be applied to possible Quake mods. So without further ado, here are the top ten Quake 1 mods I'd like to see ...

1. Team Fortress Bots
This seems to be the Holy Grail of bot mods. It would probably be the best single player mod ever, but is it possible? Is TF too complicated for bots? It has been tried but not done well ... yet. I've heard rumours that one or more designers may be working on this. And of course, the TF Bot Bot page is still active at http://www.planetquake.com/tfbot/.

2. Jailbreak Bots
Yes, JB is a Quake 2 mod, but it's still one of the hippest game ideas I've ever heard of. The gameplay would be relatively easy for bots too understand. And there is a little known Quake 1 mod called Prison Break at cdrom.com that could be used as a base. Hmm. These bots would be sweet.

3. Future vs Fantasy Bots
FvF is an oldschool mod which features an ecclectic mix of new player classes, including a ninja and a cyborb camper. It was so big it even went commercial. But instead of bots, they added new monsters ... duh! We don wan no stink monsters! Just like TF Bots, though, I've heard that somebody is now working on FvF Bots. You go boy!

4. HeadHunter Bots
Have you noticed a pattern here? The gameplay of this Q1/Q2 mod is again simple, which just invites bots. Sometimes the simplest games can be the most addictive, as well. Once more, I've heard rumblings that somebody was working on HH Bots, but that's looking shaky. Damn.

5. Quake 2 to Quake 1 Conversion
Yeah, you read right. I must be the only guy in the world who doesn't play Q2, but I'm the only guy who seems to like it. The models are superb, the art direction wonderful, and the weapons are sweet. I'd love to see that railgun or that buffed player model in Quake 1. I'm drooling now.

6. Killer Quake Pak 2
I don't mean this literally; I just wish someone would gather up some more weapons, new deathmatch modes (like, um, Paintball?), throw in a bot, and release a great new combo mod. Cdrom.com is just full of diverse patches just ripe for picking. Get to work.

7. Your Bot
You know I had to add this one. If you have visited or frequent this site, then I also know that you're interested in game design. So why haven't I seen your bot, huh? I'm waiting.

8. Disposable Heroes
This is actually a real upcoming commercial add-on CD. It will feature several deathmatch mods, maps, new player classes, and --gasp!-- bots. That makes it the um, er, the first retail product for Quake to feature bots. That alone is significant (and kind of sad). It's mine soon.

9. Starsiege: Tribes Bots Conversion
Uh, I know, this sounds like a weird idea. I just like the idea of that game, the idea of team strategy and different team objectives. Just wish somebody would do something like this for Quake with bots.

10. Run
Chainsaws. Crunching football sounds. Penalties. Bots running everywhere. Farts. What's not to like? Uh, yeah, this is my own mod (sorry for the plug). But I still wish I could find time to finish it.

 

Will the Online Game War
Kill Online Gaming?

      There's a line in the movie Superman II, one of the best sequels ever. Near the end of the film, Superman challenges General Zod and company on the rooftops of Metropolis. A cabbie pokes his head out of his taxi, looks up, and says:
      "Man, this is gonna be good."
      And it was good, too. That's how I feel about the upcoming online gaming battle between the two super titles, Unreal Tournament, Quake III: Arena, not to mention Team Fortress 2 and Starsiege: Tribes. These games will change the way we play, both on the net and on the desktop.
      Of course, they will bring a new level of social interaction and internet convenience. That's great. But they will also bring a new era of bots. Epic Megagames and id software are including these simulated human opponents as practice for net newcomers and low-skill gamers. Basically they want to use bots to push multiplayer gaming into the mainstream.
      But that notion begs the question: if we have intelligent bots that behave realistically and understand different modes of gameplay, will we still need online gaming?

      Epic and id designers have recently talked about these titles in the excellent gaming magazine, Next Generation. They have explained that the bots will navigate well, take strategic orders, miss realistically, and possess individual personality.
      "We definitely want a handful of really distinct, unique guys," says Brandon James, a level designer at id software, of the Quake III bots. "And we're going to have a bunch of guys that are schmoes, like the Red Shirts in Star Trek for people to get in there and kill."
      Stephen Polge, Epic programmer and author of the legendary Reaperbots, has similar visions for the Unreal Tournament bots.
      "There's been improvements to the bots pretty much in every phase," he says. Polge continues by explaining that the CPU-spawned opponents will have varying combat styles, levels of alertness, and favorite weapons as well.
      Most impressively, the bots will even understand Tournament's different game styles, including Capture the Flag, Domination, and Assault.
      "I think people will play four-on-four Capture the Flag where there's just one human player on each side leading the team," Polge comments. He adds that that you can even play with all bots.
      Hmm. They sound like great opponents to me. And since they don't bitch, cheat, hack, camp, insult, lag, or hog, and since you can control their exact number and the map you want to play, they may sound like perfect foes.
      I'm not dumb enough to compare bots to humans, but honestly, in gameplay terms, I cannot see a huge discrepency. For some strange reason, I'm still not burning with the urge to jump online and play with strangers. For some strange reason, the bots sound exciting -- and sufficient -- to me, in the same way that the Metal Gear Solid genome soldiers or the Half Life bosses are.

      And of course, by Epic Megagames' own admission, internet gaming can still be a painstaking chore.
      "If you think about it," says Cliff Bleszinski, lead designer on Tournament, "to play online you've got to get a computer, get the game you want to play, get an internet connection, download Game Spy, find a good server -- and the when you finally get on there, in five minutes you get your ass kicked, and you never want to play again."
      So will gamers find themselves enjoying the bots, hour after hour, mysteriously forgetting to dial into the Tournament server? This is the undertaking that has in part kept me away from online gaming. I don't think a little practice with Quake III bots or Tournament bots will lesson the hardship.
      The bots are meant to simulate a human enemy; that is, they substitute for a human. So perhaps what these two games may do is help to create a cool new genre: simulated multiplayer game.
      Just look at two things. One: even the term "multiplayer" does not exclude bots. When you play a game with the FrogBot, it's still deathmatch, right?
      Two: if these two game companies, two of the greatest in the industry with some of the best programmers around, if they both put all their talent into simulating a multiplayer experience, won't it actually seem like one?
      These designers want to bring more people onto the net to play games. But they may end up eliminating some of the need to do so.
      In any case, man is it gonna be good.

 

Bot of the Future February 27

      Back again. Been busy this week on doing some research and development on a certain bot game. More tutorials to come soon.
      The following editorial is quite old now. I should've posted it a while ago; some of its predictions are coming true.

      I still remember the first time I played the Reaperbots. I knew right then and there that things were going to be different.
      That is, I realized that these bots were more than just some mod or patch. The Reapers were a whole new game, a whole new type of game, in fact.
      I remember playing a lot on DM3, the Bad Place. I would just watch bots on the different platforms shoot shafts of lightning downward and others fire nails back upward. I realized that Quake was nothing but an engine, a perfect engine, for multiplayer games. And the bots were right at home in it.
      Artificial intelligence will never be the same. With the Reapers, we finally had artificial opponents that used the same weapons as you, used the same skills as you, and competed for the same resources as you. They shared the game world with you, in other words.
      Hopefully, action games will never be the same as well. We are only witnessing the beginning of this phenomenon now. For instance, Unreal is the first significant retail title to include bots. Also, Quake Arena will reportedly ship with these artificial antagonists.
      So, since bots are being accepted so slowly and since they have not developed all that much since the Reapers, what kind of digital opponents will we be facing in the future? The following is a list of predictions.
      No, that's a lie. It's a wish list.

      1. Deathmatch
      Obviously, we'll still see a bunch of normal DM bots. It is an untapped market as far as retail projects are concerned. Perhaps we will see a company lease the Quake engine to develop a bot-only game.
      Here game designers will try to inject personality into the bots. They may focused on building the fear, suspense, and anger of a human deathmatch.

      2. Team mods
      This is another obvious one, since we all want good TF bots, Powerball bots, CTF bots. Even the Reapers are more fun in teamplay mode.
      However, game developers will be slow to give us this. Team tactics are simply the most difficult to simulate. Teaching the bot how to think like a medic or a sniper can be difficult. But just like above, this is an untapped field, where dozens of mods are just begging for bots.

      3. Tournament
      Now we're getting into unfamiliar territory. I see a "tournament" bot game as a combination of Quake and Mortal Kombat. It will feature a good number of digital deathmatchers, say 20 to 30, that possess individual weapons, skills, models, and personalities like the contestants in fighting games.
      On each level you will have to defeat one or several of these guys before you can progress to the next level. The emphasis would not be put on picking up the best guns, but selecting the right weapon or skill for the right opponent.

      4. Hybrid
      Some people still like regular single player mode. That's fine. But I think everyone would agree that the monsters could use some more character.
      We could easily invent a hybrid, a cross between a monster and a bot. Not just a creature that will strafe a lot, but one who will steal ammo, hide keys, even call more monsters into the map.
      This kind of foe could create opportuntities for mission-based gameplay on the spot. In other words, an ogre may raise a drawbridge, forcing you to find a boat to cross the moat. Or an cyborg may set a self-destruct device and leave the ship, forcing you to find the main computer and deactivate it.

      5. Cooperative
      It seems that cooperative play has gotten a bad rap, or at least mixed emotions. But it is the one mode which can capture the excitement of deathmatch and the continuity of single play.
      This is also the mode where John Romero and Daikatana may make their mark. In that game you will reportedly have two bot sidekicks which will help you, talk to you, make friends with you. This may add an incredible emotional aspect to action games.
      Of course coop includes team games as well, and can include bot teams too. The story of the solitary space marine responsible for single-handedly saving humanity is wearing thin. Imagine if your platoon had survived the crash at the onset of Quake 2. You may have had three or four bots to command during the game, giving it an almost Command-and-Conquer feel.

      6. Competitive
      I see this new type of play as the opposite of coop, though it is more complex than that. Again, we all know the sole marine storyline is old. So listen to this: the monsters have invaded your planet and taken control of a valuable global resource or location, throwing governments into confusion and conflict.
      Your army sends you to defeat the monsters. But another governent sends a few of its troops to do the same, since they want to control the resource when the monsters are defeated.
      The gameplay forces you to fight both monsters and bots. That is, as you try to kill the creatures and get the keys on each level, the enemy troops are trying to do the same. You may be shooting aliens, and then a marine bot rushes into the room and shoots at you.
      Meanwhile, the enemy bots in other areas of the map are stealing keys, locking doors, killing monsters, setting traps. If they complete their mission and reach the level exit before you, they win.
      Then at the end of the game, not only would you have to fight the boss but the four marine bots at the same time.

      Of course, these are but mere ideas, and we all know that the game industry doesn't revolve around ideas. I think it should be noted that all of these would be radically different gameplay styles in the Quake engine. I want to note this because even the so-called "mission-based gameplay" in Quake 2 still came down to picking off monsters.
      Skeptics may ask what the difference is between killing a monster and killing a bot. It's a good question, but it has a clear answer. A monster is there to be killed. A bot, if created well, is competing with you. He wants to beat you at what you're doing. And he's going to do his thing whether your in the same room or not.
      There is another question, however, that cannot be answered. I wonder about it every day. I hope for an answer every week. I fondly recall my memories of those Reaperbots giving each other the beatdown and scratch my head.
      The question is: if we can come up with these exciting ideas, why have we yet to see significant progress with bots?

 

To Edit Or Not To Edit

     I remember the disappointment when Quake was released.
     While I felt excited about the game, I recall reading quotes from gamers in magazines that called it Doom with polygons, a dreary environment, an unfinished project. They said Quake had no story, no flair, no attitude.
     They were probably right.
     I say that as one of the games biggest fans. But I can recognize reality when I see it. Quake was a rather plain game, with dirty colors, uninspiring weapons boring creature models.
     That, of course, was why it was such a bloody success.
     That is, it came with QuakeC, and it just begged for modification. This game was a tabula rasa, a blank slate. We gamers could fix those dumb monsters and improve those sleepy guns ourselves. An online modification community, the first of its kind, exploded.
     Now we have Quake 2. Like everything in the hype-filled 90's, it came with mixed emotions. However, it does serve as an improvement to number one. After all, the monster design and animation is superb and the weapons are simply gorgeous.
     This is a strike against it, though.

     Think about it. Think about those beautiful models and textures. What is left to modify? You've likely noticed that Quake 2 conversions are few and far between. Most amateurs cannot improve on what driven professionals have delivered.
     On top of that, QuakeC is gone in favor of Windows .DLL files, which are generated with more complicated languages. To me, this just finishes the idea that Quake 2 is a polished and professional game without the need for modification by average gamers.
     There is a saying that goes, "Be careful what you wish for, you may get it." For a year we voiced our complaints about Quake 1, and well, I guess id software listened. The sequel is colorful and exciting. But now we no longer have a blank slate on which to paint.
     It's a bummer, really. This made me realize that editing Quake 1 is better than editing its sequel. In fact, I no longer think of Quake 1 as a game. To me it's a game development tool, a construction kit, a 3D programming project. For that reason, it may never be obsolete.
     With little experience -- and perhaps little talent -- I have learned to edit Quake 1 maps, models, and programs, and it has been fairly easy to come up with something. That is why I want to recommend to you to try modifiying Quake in some way. I don't think you need to feel inhibited.
     My artificial intelligence tutorials may be a good way for you to start. Even if you've coded QuakeC before, you should take a look. With these lessons you could make smarter monsters for conversions or add bots to some cool deathmatch mod.
     Whatever you do, I hope you don't think of Quake as yesterday's game. Just think about tomorrow's mod.

 

Death To All Monsters!
Part 2

     It looked bad for the humans.
     The demons from Hell had opened up a portal between their world and ours. They brought with them the most deadly weapon in all of war: boredom.

     For years, they used it against us to ruin our 3D action games and control our minds. But now it looks as though the humans are rising up again. We have a new awareness of what is fun. We don't want any stinking monsters. At all. Not a one.
     We want deathmatch.
     I'm talking about Unreal bots. I'm talking about Quake Arena. I'm talking about eliminating the fiends from another dimension entirely, once and for all. Give us bots or give us death.
     Yes, that sounds radical, but desperate times call for desperate measures. Look at Quake 2. True, it was an improvement over the first. But it was too little, too late. The "mission based gameplay" and "improved creature intelligence" made little impact.
     After all, once you've played Capture the Flag or Team Fortress and gotten a taste of human -- or even bot -- blood, monsters seem lifeless in comparison.
     And Quake 2 taught me something about myself. I already knew I hated running around for colored keys. I learned I also disliked flipping a switch to open the floodgates, or whatever those muddled missions were. Still seemed like a tedious script of events forced on me.
     So I was very happy to hear the news of Quake Arena, very happy indeed. It will be a pure deathmatch game, with bots thrown in for the single player. I'm surprised it took someone so long to realize what we wanted in our games.
     Also, I was very shocked to hear that it was coming from id software. It was them that helped the monsters open up the inter-dimensional portal to begin with. It was id that served us up with 100 creatures per level. It was id that pushed us toward 3D acceleration. It was id who was the lords of eye candy.

     Of course they could really screw up Quake Arena. Or they could announce that they had to leave the bots out to make their holiday release date. Yeah, they've bucked things up before. They are a company to which I do not swear loyalty.
     But if they take risk, I will take it with them. If they build bots, I will come. John Carmack mentioned in his .plan that the idea of QA was a "gamble." Well, I say to him now: No it's not, not at all, not if they make the game they want to. Gambling only comes into play when you include or exclude game features because of marketing.
     I am so enthusiastic about this idea -- in its ideal form, of course -- that I think every other game company in the 3D action business is scrambling right now. If you read the computer mags, you know there is a whole fleet of next generation games on its way. I believe that every one of those is rethinking its strategy as we speak.
     After all, that is the way the business operates, by trends. That's why Unreal ended up having portals and graphics acceleration, because they became buzzwords even as the game was nearing completion. That's why Prey will be a "hardware only" game, since that is known as the Next Big Thing.
     But now id software has said, well, maybe it isn't. All those new games coming out with monsters and mission-based gameplay are still Doom-clones. The monsters won't capture your flag. They won't compete with you in points. Run, shoot, run. Blech.

     That brings me to a question that some of you may be considering: What's the diference between shooting monsters and shooting bots? Good question. I said I don't like mission-based gameplay, but isn't Team Fortress mission-based gameplay?
     Yes, in a way, it is. You can even be running around looking for a key. But TF, even with bots, forces me to implement team strategy. My gun is not my only interface with the world; I have teammates, and I have a role to play.
     But at the core, though, it is a different gameplay because I get the opportunity to plan new strategies and approach the problem in new ways each time I try to capture the flag.
     When you're trying to get a red key, its just sitting there, and if you have more guns than there are monsters in the room, you win. But with TF you can rush in alone, bring escorts, sneak in the back, even choose different classes. And naturally, the opposing team with have chosen a different defense plan: there may be ten guys there, or none.
     Monsters would never be that flexible, or smart, or interesting. That is why we will win the war against boring gameplay. These creatures have been knocking us around for years. And whether it will begin with Quake Arena or the next game inspired by it, our victory campaign has begun.
     Death to all monsters!

 

Death To All Monsters!
Part One

     I am extremely excited about the Quake Arena news. I cannot claim to have predicted a retail single player deathmatch game, but that is what I've been talking about and trying to make for the last year.
     Indeed, I had just recently published my editorial on AI acceleration vs. graphics acceleration. Then we discover that id software will not pursue aggressive graphics features in favor of a deathmatch product feeaturing bots. Coincidence? Probably. But here is an editorial I wrote a couple months ago and never published. I have changed only a bit of it. It sounds even more timely now.

     The monsters are back. The monsters from Hell, from outer space, from another dimension, from the realm of cliche.

     And this time, they have more evil plans than ever: to bore the human race to death.
     For nearly five years now, our species has been beaten down in the form of dull 3D action gameplay. The creatures confused us with generic map design. They made us weak by forcing us to run around searching for keys. These demons even controlled the minds of our greatest programmers.
     No wonder why game designers make monsters the enemies in computer games: because they are so frustratingly boring and stupid that you want to slaughter each and every one. These designers have all but ignored the development of much better opponents: bots.
     Game store shelves are lined with Quake add-on packs, yet not one of them includes a bot. The net is full of Quake conversions, yet the programmers barely touch artificial intelligence. We should eliminate monsters in games before they eliminate us consumers.
     As sadistic as this may sound, it's much more rewarding to frag a human player than some nameless creature from beyond. Admit it. That remains true even if the player is a simulated human. Think of a Reaperbot and a shambler.
     Though it may seem small, the difference may be that one actually looks like a person. Whether this tricks you into believing you're on an internet server or makes you think you're fragging your good friend Billy, it doesn't matter.
     Allow me one moment of psycho-babble to explain. For years we've battled dumb monsters who line up to get shot and for months we've fragged players who rocket jump and grab their crotch. Thus, we've learned to equate a creature model with stupidity and the human face with intelligence.
     I've spent a long time developing my Minion monster bots. These "mobots" are smarter than some bots, can do things some bots couldn't dream of. But I've spent a little time on Johnny Hunter, my deathmatch bot using a player model. And it's still more of a psychological rush to kill him than the mobots.
     That's probably why game designers put nearly a hundred monsters on one level. They try to replace quality with quantity. It doesn't work. At all.
     I played the Reapers and CTF Bots every day for months. In fact, I played each of them more than I played normal Quake. To me, there was a game called "The Reaperbots," and it was much better than Quake. The experience was that distinct. And that exciting.
     So why haven't there been any commercial bots?
     It was those damned monsters. They weren't killing the human race with an intergalactic deathray or with a division of Hell Barons or with a portal to another world. They were killing us with dull gameplay.      The only game to my knowledge that even tried to be a single player deathmatch game was "XS," a GT Interractive game last year. There were about 50 bot contestants, each with his own personality and weapons. There were about 10 levels. Unfortunately, it wasn't very playable.
     Then there was Unreal, the first leading action game to include bots. However, the monsters has already tampered with the game engine and controlled most of the game. With their demonic powers, they pushed pushed back the release date. Now the game seems like a pasted-together afterthought.
     The monsters had won another battle.
     But they have not won the war.

     To be continued . . .

 

Finally

     I normally don't post general Quake news, but I have just read the most shocking news ever. I have been waiting since Quake came out for someone to license the engine just to develop a multiplayer game, including bots.

     It has finally happened.
     Finally, someone realized what we really want. Screw monsters and keys; bring in the bots!
     You see, John Carmack announced that id is abandoning Quake 3 and making way for a new game called "Quake Arena." Check this out:

     My last two .plan updates have described efforts that were not in our original plan for quake 3, which was "quake 2 game and network technology with a new graphics engine".
     We changed our minds.
     The new product is going to be called "Quake Arena", and will consist exclusively of deathmatch style gaming (including CTF and other derivatives). The single player game will just be a progression through a ranking ladder against bot AIs. We think that can still be made an enjoyable game, but it is definately a gamble.
     In the past, we have always been designing two games at once, the single player game and the multi player game, and they often had conflicting goals.
     For instance, the client-server communications channel discouraged massive quantities of moving entities that would have been interesting in single player, while the maps and weapons designed for single player were not ideal for multiplayer. The largest conflict was just raw development time. Time spent on monsters is time not spent on player movement. Time spent on unit goals is time not spent on game rules.
     There are many wonderful gaming experiences in single player FPS, but we are choosing to leave them behind to give us a purity of focus that will let us make significant advances in the multiplayer experience.
     The emphasis will be on making every aspect as robust and high quality as possible, rather than trying to add every conceivable option anyone could want. We will not be trying to take the place of every mod ever produced, but we hope to satisfy a large part of the network gaming audience with the out of box experience.
     There is a definite effect on graphics technology decisions. Much of the positive feedback in a single player FPS is the presentation of rich visual scenes, which are often at the expense of framerate. A multiplayer level still needs to make a good first impression, but after you have seen it a hundred times, the speed of the game is more important.
     This means that there are many aggressive graphics technologies that I will not pursue because they are not apropriate to the type of game we are creating.
     The graphics engine will still be OpenGL only, with significant new features not seen anywhere before, but it will also have fallback modes to render at roughly Quake-2 quality and speed.

 

Bots Take Over The World

This is another email letter I received in response to my AI acceleration editorial. It brings up a side which I didn't consider, to say the least.

     To speak on this rather uncalled-for subject, I'd like to ask one question ... why does man need Artificial Intelligence; artificial, meaning that it can make as many mistakes it can without without worry that it will die, even if it was never alive anyway?

     If man strives to make AI perfect, doesn't that narrow us down to a lesser being? Machine over Man? Bot over gamer? If one has read the book, Frankenstein, you know already that seeking to create psychological power is the worst mistake man could make. Frankenstein could think faster than a normal man, move faster, and react faster. In a way, Victor released a superior being onto the world despite its infancy.

     Now consider a little forthought ... What is the use of an AI accelerator board? To create a standard for infinite artificial intelligence; thats what!

     Therefor, since the intelligence can now make as many mistakes as it can, and start over as many times as it can due to the flexibility of hardware over software, it can learn, and therefor become perfect by manipulating its underling, the HDD, as a memory component. Much as the Eraser bot for Quake2 recaculates every incorrect pathway through a level until all items are reachable from one, solid path, so that it can become a perfect opponent.

     An AI accelerator, under certain programming, redoes every solution that can posupset its only enemy: Man, the creator.This is possibly the worst mistake man could make. AI can duplicate hardware to software in seconds, much like video game cartridges are converted to rom DAT format.

     I say stick to software AI. Polge was right in some awkward aspect. Software CAN duplicate itself(virus), but it has a priveledge that cannot be denied; PURE DELETION.

     If you're gonna make AI for fun and games, do it, and keep it simple; but stay away from the hardware. There are people out there who are waiting for something like this to be made, so that machine can finally upset man.

     I rest my case.
The Guard Dog

 

A Response To AI Accelerators

Coffee,
Neat idea. As a non-gl-having player, I've yet to see the light as to why colored lighting and other nonessential eye candy has to take priority over AI. I'd love to have one, if they weren't too expensive.

With that said, I don't think you'll ever see an AI accelerator. For one, the market just won't be there if you just target the gaming crowd. The non-gaming market will want to know why the processor itself can't handle it...and why a second processor on the motherboard wouldn't do the job.

Another issue will be standards. Look at the nightmare that was the V.90 standard and what DVD is going through right now. You think the industry could agree on something as complicated as an AI standard?

Of course, if you try to generalize everything then you've just got a souped-up second processor on a daughterboard...in which case a dual-processor system might be a better solution anyway.

Or I could be smoking crack. Interesting idea, nonetheless.
regards,
Rantage
Clan Steel Maelstrom

 

And God said,
"Let there be (colored) light"

     Actually, God never said that.

     But don't let the game companies and hardware makers here you say that. They're all into their colored lighting these days. Soon we'll have pink kitchens and purple bathrooms in games.
     I could have called this editorial "I have seen the (colored) light," for recently I downloaded a gl driver to try Quake 2 in gl mode.
     I started it up. A day later, I saw it.
     Yeah, it loaded forever. Windows 95 must have filled all the empty space on my hard drive with orange outhouses. Next time I'll try a Pentium 7000 with 424 meg of ram.
     Anyway, I think my shotgun was giving off polka-dotted light. I touched the mouse, and it began to load again. The next day, when it stopped, I turned it off.
     Needless to say, it wasn't a religious experience. Yet Quake 2 is highly identified by colored lighting and other similar graphics effects. Its creature intelligence and originality is lacking, though.
     I though that Unreal was going to be different. Good AI, good story. But it was the Unreal team themselves who halted the game a year ago to add portals, colored lighting and other similar graphics effects to the engine.
     In the end, id Software and Epic MegaGames made the same mistake, confusing their priorities. But you know, I have nothing against 3D-accelerated games. Both of these titles are visual orgasms. They're just not that stimulating when it comes to gameplay.
     So why don't we have artificial intelligence accelerators?
     Think about it. AI takes up a lot of CPU time. Lots of formulas need to be run even for a bot to run across a map. You couldn't play 100 Reaperbots on a normal system.
     There could be a microchip that handles just the math of AI, so the engine wouldn't ever lag. Bots could not only be much faster but much smarter (indeed, I read that Stephen Polge, Unreal AI author, said he couldn't do the true AI he wanted due to current CPU speeds).
     I could live with 3D acceleration if we had AI acceleration. Meaning, I don't care if you pander to the I-want-eye-candy-crowd as long as you cater to the I-want-AI-related-gameplay audience.
     Then you could have sparkling yellow light shining down on a creature who has just stole your keys and locked all the doors in the level. Or rich purple light illuminating a monster who just challenged you to a game of Capture the Flag.
     That is the game I want to play. Call me crazy.

 

Coffee finally announces
new project:
The Baywatch
Quake Conversion!

     Oh yes, this is going to be a great one, the Baywatch Quake Conversion. It's about time someone had the nerve to do a project like this: a mix of sex, violence, and, eh, more sex.

     You've played a hardened space marine, but are you ready to be a lifeguard?
     In Baywatch Quake, you can choose from three drop-dead gorgeous characters. First, there's C.J., with her breast-mounted rocket launchers. Next there is Summer; whoever looks at her cut-off jeans drops over from a heart attack. Last but not there is Mitch, with his killer-sharp hairdo.
     Of course, your enemies will be just as deadly. You'll battle evil beachcombers, purse-snatchers, and psychotic ex-girlfriends. We would do ground-breaking artificial intelligence, but then we wouldn't make the deadline.
     The scenery will be just as unbelievable. There will be a City map. There will be a Beach map (yes, sand is hard to do in Quake, but our map designers have done well). Then there will be an Underwater map, my favorite, where most of the drowning takes place.
     But the goal in Baywatch Quake isn't to save victims, oh no. Nor will your goal be to run up and down the beach making sure your jiggly body parts are jiggling and your tight body parts remain tight.
     No, actually, there really aren't any new goals. Thinking of new gameplay really isn't our job.
     But like all other Quake Conversions, this will have 27 new models, 35 maps, and 120 new sounds. Don't ask us why, that's just what we do.
     I've been foolish to work on dumb games like Minion and Minion Arena. Who wants unusual gameplay? Who wants different AI? No, I'm going to invest all my future time in making the file size as large as possible. We anticipate that Baywatch will have nine files to download, each three meg each. And after you see all of that stuff, you really won't believe you downloaded it.
     Baywatch Quake. A Whole New Dose of The Same Old Thing.

 

By the way, what do they call April 1st?

 

Good Games,
Bad Games, Part 3

     Critics say that we Quake fans are just a hormonal bunch of thugs who want to shoot things to death. Well, let's face it, there's nothing like watching those spinning Perforator barrels fire nine-inch nails into some slashing fiend.
     But we have to want more than this. It gets kind of shallow after a while. Playing even the best Quake conversion is like eating Chinese food. It tastes oh so good but leaves you empty.
     We need action *and* adventure.
     Here we reach a thin line. A Quake mod doesn't need to test your IQ. It doesn't have to make you a smarter person. Personally, even though I am near 30 years old, I don't like more cerebral games like Ultima and Civilization. I want instant gratification.
     Yet I want that feeling of gratification to stay with me for hours, days, or weeks as I play the game. I want to get to know helpful characters. I want to hate evil villains. I want to use my rocket launcher to blow a new exit into the map.
     Speaking of that, I also want to see the end of level-based gameplay. That has got to be the one aspect ruining everything. There should be something so fun on each map that you don't have to go to the next one.
     If I'm being a little vague on that, think of this: if it were up to me, I'd get about 30 Team Fortress maps together, throw in some bots who understand that deathmatch game, and release that as Quake 3.
     Or how about releasing something like "Quake Olympics," where you have a handful of similar bot sports, and you have to qualify in one to go onto the next.
     Well, you might not like sports. Okay, say you have one huge continuous Quake world full of villages, cities, and castles. As you kill the monsters in each of these regions, you become the leader of the region. You get to go back for its weapons, ammo, treasure, food, helper bots, and more.
     I know these concepts are a little raw, but I'm literally thinking of them off the top of my head. And that just goes to prove that given a little thought, you can come up with something more innovative than the current single playerQuake mods.
     Imagine what would happen with a lot of thought.

 

Good Games,
Bad Games, Part 2

     You know what gets me: all these "conversions." I always thought that was a silly word and a silly concept. To me, if you are doing a conversion it means you are taking someone else's idea just to have a project.

     We scold id software for rehashing old gameplay and even old engines, yet we go out and make "Some Other Game" Quake or "Not Really My Idea" Quake or "I Was a Console Game in 1982" Quake. A person can't get home by walking backwards. Think about it.
     And people choose some downright weird topics for conversions. I mean, if you want to people frustrated at the state of unoriginal Quake mods, just go to cdrom.com. Oh damn, that's like the home for everyone who wants their name in a readme.txt file yet who has no originality whatsoever.
     Just look at "Syndicate Wars" Quake. It's like, five new weapon models. No, really, that's it. No missions, no story, nothing. And how about "Command and Conquer" Quake. That's about five new weapon skins. Not even models. Argh.
     Of course there are many projects, which brings me to the other extreme: the total conversion with 190 new weapons and 77 new monsters and 342 new maps. Except that the file size is 54 meg or something. And after you spend three months downloading it, you get to take the key and exit the level. Ugh.
     Yeah, the single player who likes Quake just doesn't have much to choose from. We're screwed. You'd think if we could execute blended colored lighting, extrapolated frame animation, that we could come up with something to play.
     Yet there have to be some fresh, exciting things we can do for the single Quake gamer. I'm sure even you have some ideas. Everyone does. We'll discuss those next time.

 

Good Games,
Bad Games, Part 1

     This is the first of three parts of an essay I wrote at work. It's a discussion of the state of gaming now. It'll tell you where I'm coming from where games are concerned.

     Single players are still screwed.
     When I started this website last year, I founded in with the Quake single player in mind. At that time, there were some great net-based deathmatch sports like SWAT Team, Quake Rally, and Team Fortress. Yet there were almost none like that for gamers with retarded modems.
     It hasn't changed a damn bit.
     I mean, there aren't any more deathmatch games for us now either. Oh yes, there are a lot of *mods* for single players. Mods with a bunch of shiny new textures and monsters with more teeth than IQ points. Mods that offer you a hundred new maps if you download their 25 meg file.
     And yes, there are quite a few bots out there. Bots that swim efficiently and move realistically and bore you to death effectively. Every bot is argued to be smarter than the other, but who cares. They don't feel fun.
     So where are the deathmatch games for the single gamer? Now don't look at me like that is some kind of contradiction in terms. It's not. Capture the Flag Bots is a DM game for an SP gamer. Unfortunately, it is the *only* one I know of. And it was released a year ago.
     I may be more talk than action at this point, but I do consider Fugitives in Fragtown and Minion Arena to be single player deathmatch games, as well as the upcoming Minion.
     It's also true that I don't always get to update the page as often as I'd like. But I've recently checked out some projects that began or were around when I started, like CTF Bots, New Union and others, and they haven't updated all. I wish they would.
     My point is that even though I may sometimes slip away, I haven't left. Nor will I any time soon. I've been programming at home for about 10 years now for only one reason: I want to see good games get made.