Darwinian Diversion

After beating my brains for a week, I finally reached the point on Saturday that I looked at what I created but realized that I was running on empty.   While the code may have been a bit better built through practice, the game I created was even less interesting than my previous versions.  My “keep it simple” focus may have made me a better coder, but I was a significantly worse game designer than I was just a week ago.

Clearly, my batteries needed recharging.

I had been waiting for Darwinia to become packaged with Multiwinia for a good value, and discovered that Steam was selling the two packaged for $15.  I’m glad I shelled out for it.  The novelty in the intros alone was worth the entrance price, the rest of Darwinia was riveting enough that I completed the main storyline and all the few side missions in one sitting (about 10 hours).

Darwinia is remarkable in that it is very much a dynamic world game.   Through your actions, a virus-infested digital world with a broken infrastructure becomes mended.  The changes you make will persist even if you go to other maps – a few viruses might inexplicably return, but if you built a powerful defensive infrastructure, your Darwinians can now handle themselves.   Leave a map and be taken to a map selection screen where apparently the little digital souls – the consequences of your warfare – can be seen floating through the virtual cosmos between maps.

Brilliant, brilliant game.  Well, the interface was a bit awkward here and there – using explosives with my squads was as likely to kill them as it was the enemy because I may not have been able to notice the curvature of the terrain – but nonetheless Darwinia was a thoroughly enjoyable experience with a deep message throughout.

Consider my batteries recharged.

It seems that my “keep it simple” focus in game development was regrettably misplaced.  The purpose of coding is to create code which serves your needs, and a “keep it simple” focus was me attempting to serve the code’s needs.  What I really need to do – and I’m bad at getting myself to do this – is sit down and brainstorm the awesome game concept first, working out just enough essential details that I know how to make it work.

The Drawing Board Never Rests

Thanks to my “keep it simple” focus, I have finalized a somewhat powerful mechanism of open-ended interaction in my BYOND game, which is indeed turning out to be quite the Roguelike.  I’ve successfully streamlined the concepts of traditional inventory management with a mechanism that provides a much more direct conduit to providing the players with a wide variety of choices of what to do in a role-playing situation.  However, I’ve yet to finalize a rather important aspect of my design, that being how the game is played or even the overall flow of it.

I’ve figured out these things a few times in the recent past, but I’ve broken my previous molds when I went for a “keep it simple” coding focus, and now I have to take a pause from coding once again to figure out a whole new model that suits this new focus.  My previous designs remain littered with concepts that require multiple interface modes and other things that really are not feasible in a first design or in ways that suit BYOND’s strengths.

Although I’ve learned quite a bit about the kind of game I should be making, my overall goals remain the same as they were before:

• Dynamic content – The world is very mutable with the players’ actions.  When you wipe out a batch of monsters, they stay dead until conditions dictate they should logically return.  The rise and fall of player-made cities.  Things like that.
• Interactivity – Players should be presented with a good number of meaningful choices to make.  “Attack” or “Use Healing Potion” isn’t enough.  I’d like to have a situational system where a conflict rarely plays out the same way twice.

Despite knowing the difficulty behind reinventing the wheel, my goals will require that I do so to some extent.  Little wonder it’s taking some time to see this to fruition.  I suppose it’s just as well I’m not keeping things completely simple: I’d like to have some novelty to my game.

My school responsibilities have vanished off the radar.  I’m going to have a lot to answer to come Monday.

Maybe

My BYOND work continues pretty much as I said it would be just a few days ago.  If any new development has come about in my BYOND work, it’s been that I’ve been taking it back to the drawing board for yet more complete refinements.  As I said before, it’s a cycle thing: design, code, reconsider, over and over again, each cycle improving the result.

Currently, my focus has changed to simplicity.  Very ambitious games require a lot of arbitrary code be written to support them: my Mass Effect-like design would essentially require coding three separate games and tying them together.  I’ve reeled back my ambition a bit and realized that, while BYOND is nice in that it offers enough flexibility to realize a considerable range of dream games, you can realize a finished result much sooner if you settle for reinterpreting your dream in BYOND’s terms.

But I really didn’t start writing this blog entry with my BYOND progress (as little as it has been) in mind.

Fallout 3 Construction Kit & Downloadable Content On Its Way

Well, shut my mouth, I suspected but I didn’t think they were actually going to do it: Bethesda is releasing a construction set for Fallout 3 after all.  Perhaps they were always planning on doing this, or perhaps the fact that the community has largely hacked the game to produce their own mods forced them to go ahead with it.  In any case, the fittingly-named Garden of Eden Construction Kit should be a lot of fun to play with just as the Elder Scrolls Construction Set was.

I still hold to what I said earlier that the punch-out ending would be difficult to mend without Ron Pearlman on speeddail, as the main problem with the ending was that the recap was so unsatisfying.  Perhaps some of the downloadable content will be an extended ending.  I’d pay for that (though such a thing as a correct finish to the game should really be gratis).

Beating the brains for a week

Thanksgiving Vacation is here: I’ve 8 days off, and with them an opportunity to do some serious BYOND game development.

For me, it seems that Original Game Development is very much a cyclical thing: I decide on what I want to do, code a bit, test it, and then decide on what I want to do again.  I make so little progress that, to the casual observer, it may seem I’m just really bad at it. Thus, it was with some relief that when I was thumbing through Jesse Schnell’s masterpiece not too long ago that I read that professional game designers have a similar cycle of refinement: design, code, test.  Over and over again, each cycle improving the game, with no real end other than an unmovable deadline or depleted budget, forcing them to release what they got.  It seems I’m a natural that I ended up doing this by accident.

I’ve mentioned a few times in the past that I enjoy thinking, so I have to admit, this mental exercise is a lot of fun for me.  Though you may not think highly of game developers, they really are fantastic thought technicians, not content with boring reality and so they go off and invent whole new interesting simulations of reality refined to such a minute detail that even an overgrown calculator can reproduce them.  Richard Garriott may have always been an astronaut of sorts, as what is a game designer if not a person who is not quite content with life on Earth so they undergo an incredible ordeal on a slim chance they may find a more entertaining alternative?

However, as much as I enjoy the thought exercise, there’s more at stake than a satisfying puzzle.  As far as society is concerned, the difference between a genius and a moron has a lot to do with whether or not they produce something of worth to them.  It’s not true – genius exists whether or not there’s artifacts to prove it – but society in general only cares about one thing: results.

My life is most definitively without results, and I’ve only my own love of games to blame.  If I want to prove myself was more than a moron to a self-centered society, I need to produce something to show it.  What better idea than to leverage my own love of games, dislike of poor games, and the infinite game-of-making-a-game of game design, to try and create better ones?

This fundamental line of reasoning is the main thing that has generated the focus needed to dedicate so much time towards game development lately.  It’s time to start showing the world what I can do.   If “what I can do” amounts to “suck,” well, at least it was good practice.  Maybe I’ll produce excellence if I put the game through some additional cycles, or maybe I’ll go completely back to the drawing board wiser for the experience.  Were it not for this attitude, likely I would be too paralyzed by fear of failure to do anything.  Now, I only need two things: the opportunity and focus to keep trying.

Our Douchebag Economy

Hurray, Thanksgiving Vacation is just around the corner.  If we can still eat as a result of this tanking economy, we’ll have something to be Thankful for.

I’d like to write a little about that, but I’m not sure what I’d say.  What I want to say is something along the lines of, “Douchebaggery: You Reap What You Sow.”

Making the greatest game that ever was… or not.

Partly out of the sheer heavy practice I’ve been dedicating towards BYOND lately, my project has taken an ambitious turn.  The other day, I came under the realization that if I can make a suitcase you can drop which releases a bunch of nanites which then assemble a base (and I did) then I can pretty much make anything.

The fundamental truth of the matter is that you can do anything in BYOND you want to do — well, anything involving a tile-based engine and reasonable technical limitations, that is.   Consequently, all my previous self-imposed limitations have fallen away, and I’ve become that much closer to really building something that is my own net dream.

Fixing Fallout 3, Left4Dead, and BYOND

How to Fix Fallout 3

It’s interesting how many hits I received on how much Fallout 3’s ending sucks.   I thought my Warhammer Online PvP Guide was popular, and part of that was because I bandied that about some forums.  However, I’m getting three times the hits from Google searches specifically looking for “Fallout 3 ending sucks/sucked” and I didn’t tell anyone about my post.

To Bethesda, I say this: act now while your game is still on the radar.   It’s not that Fallout 3 is a bad game, it’s that Fallout 3 is a great game that becomes a complete let down at the end.  Sure, you’ll move Fallout 4 and the expansions even if you leave this problem unfixed, but I bet you would move substantially more if you made the time to finish Fallout 3 properly.

What would be the right way to end Fallout 3?  In reflecting on the earlier Fallouts, I can think of a simple answer: create some longer recaps that talk about what happens over the future as a result of your character’s actions in greater depth.   Why this is important is because those recaps served an important dual purpose.

• First, they provide a powerful coping mechanism with being forced to retire the character.  A lengthy dialogue explaining the legacy left by your character is a satisfying payoff.
• Second, it creates a meaningful purpose to replay the game so that you can better observe the results of your actions.  So I failed to save this city?  Well, lets play again and this time we’ll create even a better legacy.

See now why cutting the recap short was shooting the game in the foot?  It also wouldn’t hurt if it’s better explained that it wouldn’t be any different if Fawkes is used than if Sarah is.

A lot of people are hoping Bethesda will release a construction set for Fallout 3, but there’s been no official announcement that one is forthcoming.  It’s just as well, unless somebody has Ron Pearlman on speed dial, any fan-made Fallout 3 recap would end up feeling something like this.  Besides, if it doesn’t come from the official source, most people won’t even hear of it.  Bethesda has to do it, and they have to do it right.

Left4Dead Anticipation

Armed And Ready

They say the essence of suffering is attachment, so lets move from the Fallout 3 tragedy for now.  My next fixation (and hopefully one with a happier ending) is Valve’s Left4Dead, the second game I’ve bought for months (the first being Fallout 3) and which is due to unlock in about 24 hours (Tuesday early morning).

Why am I so excited about Left4Dead?  Well, I’ve discussed at length the coolness of the “Virtual GM” aspects.  This game is less a Half Life 2 total conversion and more an outgrowth of the concept of dynamic games in the mold of the Roguelike with a dash of Hollywood cinematic expertise thrown in.

Despite the largely static maps, here is a game you can play hundreds of times and never encounter the exact same game twice, and it’s riveting: each element of the game – whether it’s map, props, characters, or audio – are extremely well thought out and implemented.

The retail version expands the initial 2 maps in the demo to 4 “movies” (chapters) of 5 maps each, adds additional weaponry, and more.  The fifth map of each movie is a humongous boss fight.   In addition, there is a versus mode that allows players to play powerful infected characters, essentially adding an additional level of intelligent adaptive adversaries to the environment.

That’s worth $45 of largely-borrowed money to me!  As a discerning gamer on the cusp of becoming a game developer myself, this will probably be my last purchase this year.  Well, unless Heavy Duty comes out this holiday season, and the fleeting press coverage we get from that game leads one to suspect it’s had one foot in the grave for years.  Ah, it always hurts when the good ones get away like that.

More Productive Procrastination

Speaking of which, though this week has been perforated by frequent and annoying interruptions (e.g. my duly recognized responsibilities as a financial aid student) I was able to spend much of Saturday putting working into my BYOND game and all the good concepts for BYOND games that get away.

To an extent, my progress is very much a struggle with writer’s block, but I’m learning to cope with that.  The trouble is that I want to make the game perfect the very first run, but there’s too little of it done yet to understand what needs improvement.  A solution is to say that this is fine: one of the cures for writer’s block is to forge ahead anyway, spamming out crap and being willing to throw away a ton of work if I decide I’m going in the wrong direction.

For The Win

Like anything worth learning, practicing it is hard work that leads to personal growth.  It’s why, despite looking forward to playing Left4Dead on Tuesday, the very best thing I have going for me is this mess of a game I’m developing in BYOND.  It’s why I haven’t resubscribed to Warhammer Online even though my Fallout 3 fixation is largely stunned and confused from that wet mackerel slap of an ending.

Seriously, I think I’m learning more useful stuff just screwing around in BYOND than I am with my several-thousand-a-quarter University education.  It’s part of the reason I have such a hard time paying attention to classes lately.   It’s a pity the job market sucks right now, or I’d take some reasonably-well-paying part-time gig (just enough to pay bills and afford the occasional needed hardware/software) while I hammer away at becoming good enough at making games to sell them.  (Or at least good enough at making games to understand where I can find classes I can care more about.)

Left4Dead: The Virtual GM Goes Mainstream

A concept I came up with awhile back (and I’m not saying Valve copied me or even know who I am) is the idea of, instead of having a Roleplaying Game made up of static quests that are played out, you make a game where the computer behaves as a virtual Game Master who works like a real GM.

What does a real GM do that’s so different from what we commonly see in games?

Well, if they’re a good game master, they bring both their imagination and powers of observation to the table with an understanding that the goal is to entertain the players.

In terms of imagination, rather than just reading through a pre-made scenario and doing everything by the book, you instead put the players in a setting, perhaps provide a little background story for impetus, and ask the players: what now? Then, the players can take the story whichever direction they want to go. This is the power of dynamic content.

So far, so good, but many MMORPGs have that and it isn’t enough by itself. The world may be there, and the players may be free to wander it, but this does not inherently make it an adventure.  Indeed, many a veteran player simply gets bored, learns where the experience points and gold is, and grinds that monotonously in a tried and true, safe manner.

For it to become an adventure, the world needs to provide “risky, dangerous and uncertain experiences” (wikipedia). Here’s where the GM’s powers of observation enter the picture. They observe their players and judge for themselves when inserting these experiences is appropriate and the best way to go about it.  What’s more, the GM makes sure the adventure sees an entertaining result – adventures should be surmountable if the players are doing their best – going so far as to fudging a roll if it would otherwise lead to an unjustified change.

That a game even allows the player to bore themselves is a real shame.  So, when that veteran player thinks he found a tried-and-true, safe method to grind levels, the good GM forces him to think again.  A particularly uncooperative or lazy player will likely discover their characters bearing the brunt of a fully-justified deathtrap.

So what’s this about the “Virtual” GM going mainstream?

All computer games could be argued to have a Virtual GM who runs a scenario.   An early example of a Virtual GM would probably be a Roguelike game, which randomly generates the maps and monsters.  However, that would be a very poor GM indeed.  Only a few of them provide much freedom, and slim to none actually watch the player and endeavor to create adventure.  Most CRPGs don’t even go as far as a Roguelike game in terms of providing an unpredictable experience.

Left4Dead features a “dynamic system for game dramatics, pacing, and difficulty called the Director” (wikipedia). This basically is the next level of Virtual GM, an artificial intelligence that randomizes parts of the scenario, observes the players, and tries to figure out what adjustments need to be done when they happen.

Playing a bit of the demo, I can see the Virtual GM/Director in action.

• Once in awhile, a large surge of zombies hits the players: the director has decided the players need a bit more action. If the players seem to be zombie-killing experts, more zombies are thrown at them, and perhaps ammo is provided readily.
• The goal of the scenario is that the players work together, so when one wanders of, the director doles out the appropriate punishment: one of the super-powerful zombies which require player interaction to survive show up and attack the stray.
• If everybody stops to heal themselves at once, leaving themselves foolishly unready, don’t be too surprised if the Director decides it’s time for a zombie attack.
• If the player characters are badly injured, but the players seem to be cooperating well, perhaps they’ll be rewarded with a cache of pain pills.

All of this could happen… or not: the element of unpredictability must be maintained for it seem adventurous.  I have to say, the result is pretty freaking fabulous.  I could probably replay the same two maps in the demo several dozen times without being bored of them. Every play through is different, unpredictable, dangerous, and ultimately: a high-quality virtual adventure.  Would that I could say the same thing about the average MMORPG map!

Speaking of which, another game I noticed is DC Universe Online, being developed by Sony Online Entertainment.  According to a PC Gamer magazine article I read, they’re going to have no static quests you can come across, no enemy spawns laying around waiting to get attacked.   Instead, they feature a “dynamic encounter system” that has adversaries appearing in reaction to the players, and quests manifesting as calls coming over a communicator.  It sounds a great deal like the earmarks of a robust Virtual GM implementation.

As for a third game that does this, well, that’s one of the features I’m working on with my own game using BYOND. A compelling Virtual GM is something I’ve been wanting to see in CRPGs for a long time, and it seems my procrastination has allowed Valve to beat me to the punch.  They put a lot of thought into it, and I see I’ve awhile to go before my implementation is completely fleshed out.  (Currently, I’m mostly working on revamping my code to be more modular right now, dragging my feet to get around to that vital design process.)

There’s still a lot of work to be done in terms of what a Virtual GM would be capable of.  A good chess AI can best the masters.  So, can a good virtual GM AI produce an adventure that rivals the best GMs as well?

Grinding Gears In The Idea Factory

I know I’m a fellow who enjoys thinking, but I often like to do something as well, even if that something is program or play a game. As resolved as I am to see a BYOND project through, I regret to say that a lot of my time has been spent thinking in the earliest levels of refinement. Going, “hmm,” and laying back on my chair, wracking my brain, with very little contact occurring with with the keyboard.

A brief history of what didn’t happen.

Project Xenoverse began as an idea of being a virtual world where dynamic change occurs through the actions of the players on the outside – something many MMORPGs are afraid of – while, on the inside, featuring enjoyable gameplay – something many MMORPGs sacrifice in lieu of a grind far longer than it deserves to be.

As wonderful as a seed idea as that was, it’s not very specific. Not too long ago, I took Project Xenoverse back to the drawing board because I came to realize that the lack of story and overly flexible tool given to the player left the whole thing feeling flat.

I ended up with Project Xenoverse version 2, which ultimately mutated into a bit of a AutoDuel/Battlemech game while still allowing the players to freely modify the world. Equipment that the players were used become a lot more poignant when it made up the parts of combat vehicles.

That would have been good too, but it still wasn’t quite good enough because the trouble with vehicular games is you lose that personal touch when you = vehicle on the screen versus you = person on the screen.

Yesterday, Project Xenoverse mutated into Planet Xenophobe, which attempted to bridge the gap between you = vehicle and you = person by suggesting that you = person who is a vehicle via a transhumanist premise. Between this and some other background aspects that had been floating around but had yet to be nailed down, it was a quantum leap forward in terms of plot, but it still wasn’t quite where I wanted to be.

Today, the vehicular angle is gone. Much of my teen years were spent enjoying Battletech MU*, and I may well see that through some day, but my first BYOND game is going to be more of a straight-up RPG.

Finding the why of it

Currently, it seems I’m looking at something very similar to Space Station 13, in that you are a space colonist trying to keep your colony running. Although, my design spares a lot more detail to the game aspects, adding things like advancement skills, while keeping that dynamic world focus, and is built to be a virtual world that can stay up without reset.

Even that isn’t quite enough, I don’t feel I’ve adequately established the “why” of it. Why is this colony out in the middle of nowhere suffering disasters? Why should you care? It’s a problem in which even Space Station 13 has trouble answering, as it seems many players would rather just sabotage the space station, whether or they’re running a “traitor-mode” game, or even if it was their turn to be the traitor. If I could establish the “why,” the wheels spin much better, the players’ motivations are better refined and ready to go.

I’m an extremely tough audience, and whether this proves an impassible barrier or an incredible lens which generates games the likes the world has ever seen is possibly a matter of how much will I’m able to direct at this.

Some excellent answers in a book.

I picked up The Art of Game Design by Jesse Schell awhile back, but I haven’t read it very much. It’s not a bad book, quite the opposite, I just haven’t read much of anything that hasn’t been on the Internet or related to a class for months.

Today, to break my stalemate, I gave Chapters 6 and 7 a skim, and it was very well used time indeed. I came away with two noteworthy points: First, as part of Chapter 6, I was advised to seek inspiration on how to make better games from outside of games, as therein lay a lens of infinite inspiration. Second, as part of Chapter 7, I was made aware that game development is very much a matter of a looping principle where they become better with each major pass applied.

Perhaps I’m on schedule to produce a great game, and didn’t realize it. Maybe Project Xenoverse, Project Xenoverse Version 2, and Planet Xenophobe were just earlier loops of refinement.

A new focus arises in my game development goal, and that is to code in such a way as to leave myself open to new possibilities, so that I can move on with the next loop with little difficulty. The modular programming practices discussed in the previous Blog entry will no doubt help with this. This way, even if I end up scrapping a hundred ideas, I’ll be steadily building a code base that will rapidly accelerate me to my next one.

Learning BYOND #11: The Power Of Modular Programming

Though I think of it as being a couple weeks off, this is in mostly counting minor dabbling: the actual amount of time I spent away from hardcore BYOND development is more like a month.  I’ve spent the pretty much the entire weekend (in which I had four days) hammering BYOND, and I think I’ve finally caught up with my lost knowledge.  This, and more.

I’ve picked up and dusted off my old Learning BYOND series for a special installment about the power of modular programming.