Man, that felt good. Anyway, here’s my new location – the Cave. This will only be available in the full version. You hear that, backers? You’ll be getting your first exclusive content very soon…

The castle, mine and wreck illustrate the general idea: objects you create can be locations that you enter and explore. The plan was always to have more such objects, but the Unity engine really doesn’t make it easy to have persistent levels that you can move back and forth between. In the demo, all four locations are packed into a single Unity level – which causes bugs (such as the disappearing skeletons in the wreck) and also performance issues. When I sort out a better way to organize levels, the way is open for locations that lead into more locations, and so on.

So if you create a City, it probably ought to contain a bunch of different buildings – such as a palace, a market, a jail, a bank, etc. Furthermore, the bank would also contain a vault, the jail would contain cells, and so on.

Another thing that just occurred to me: people keep telling me this game has huge replay value, and they enjoy playing just the same demo level again and again, trying to see what words they can make and how efficiently they can play. To celebrate the game’s replayability, I really ought to be putting in a scoring system.

I think, each time you spell a word, you should receive points equal to the square of the number of letters in the word. Everyone likes spelling long words, right?

However, I don’t want to push people into dull grindy “spell as many words as possible” gameplay, so here’s the plan: for a word to score any points, it must contain a letter you have never used before.

In other words: at most, you’ll get 26 opportunities to score during the game. Use them wisely!
(Ok, ok, maybe other actions will also score you points. Slaying the dragon, cracking the bank vault, escaping the jail… we’ll see how it goes.)

So how does this apply to Wordsmith? Most puzzles in the Wordsmith demo provide you with three rewards:

A) Satisfaction. (you solved the puzzle!)

B) New locations. (most of the puzzles are about getting past some kind of spatial barrier)

C) New resources. (in each new location there are usually new creatures to kill with letters you haven’t got yet; in a couple of cases, just figuring out how to kill a creature is the puzzle.)

So yeah, I’m not as ascetic as Jonathan Blow; in Braid the only reward is A. (And maybe a tiny bit of B.)

More importantly, though, I’m concerned about C. Playing through the demo, you solve puzzles to get access to new letters, and use the new letters to solve the next puzzle, and so on. It’s not sustainable like this: the alphabet only has 26 letters in it. Just letting it run out, and leaving A and B as the remaining rewards is not practical either – when the player has grown to expect C, they’re going to be frustrated if I suddenly stop giving it out. (I guess WoW’s level cap is an example of that actually being done in the wild. It might work, I guess, but for now I’m going to steer clear.)

Fortunately, I think there are a couple more tricks I can try to carry on rewarding the player in a longer-form game. The first idea is a limited supply of wildcard letters – I could make certain letters (I’m thinking U and J) completely unobtainable… except that completing a “boss” puzzle earns you a wildcard. This is a one-use, non-refundable tile that can be used as any letter at all. A little taste of infinity. Use it wisely!

The second is “golden” letters – completing certain special puzzles (or completing them in a special way, perhaps) could earn you an endless supply of a particular letter. This could be rather game-breaking, so I think this would have to be kept as just an end-game completionist’s reward. Maybe the game world is divided in some way into 26 districts, and each district has a particular “gold letter” associated with it which you can obtain in some highly non-obvious way.

That’s all I have for today…

Making a bigger world is the obvious approach. I could just add more levels, with more puzzles and more stuff. For example, when you get past the spiders in the mine, maybe the tunnel continues. Rocks and walls block your way, gas pockets threaten to explode or suffocate you, and there are fiercer and fiercer subterranean monsters to fight. Similarly, when you get past the mountain, maybe there’s a glacier to explore (warm clothing necessary), and when you get past the island, maybe there are shark-infested waters blocking your way, followed by marshland filled with mosquitoes, alligators and slimes, leading up to a jungle temple filled with traps… (Yeah, I’ve been thinking about this stuff a lot.)

Problem: as the world gets larger and larger, it gets more and more annoying to travel around harvesting letters. “Oh, I know what I need to create! – But, I need a K. Back to the wreck, let’s kill some more skeletons. (cue 15 minutes of walking back through the mine to the start position, killing a bunch of skeletons, then walking back down to where your puzzle was.)”

To prevent the game from getting too annoying, I think the world needs to remain fairly small.

An alternative I just thought of while writing: fast-travel! Such as Torchlight’s Town Portals, perhaps, or a “warp to any location you’ve visited” mechanic like the recent Elder Scrolls games.

Hmm… that does actually sound like it has potential. Wow, Paul Graham is right. Writing essays is awesome.

Anyway, just a quick comment to clarify that no, I’m not dead, and neither is Wordsmith. What actually happened is that I got stuck.

The design I used for the demo clearly can’t sustain itself across a longer game: there are only 26 letters in the alphabet, and by the end of the demo I’ve already given the player about 23 of them. If the game is about solving puzzles to acquire new letters, and using them to solve more puzzles… then by the end of the demo, the game is almost over. How do I make a longer-form game around the Wordsmith concept?

So I’ve spent a few months now trying to answer that question. The obvious answer seemed to be a Minecraft-style survival mode: try to survive monster attacks by collecting letters and making weapons.

This idea did not work. The problem is, Wordsmith is inherently a very tough game; it takes time to think of good words to make, and even then the process of harvesting the letters isn’t trivial. When you add the additional pressure of coming under attack and needing to create something quickly, “tough” turns into “impossible”. It’s like adding a time limit to a puzzle in an adventure game.

As the creator of the game, naturally I know which weapons and pets are supposed to be good for fighting which enemies – but even I was getting stuck when a new enemy showed up. It was just too hard to be playable.

I think I have a new plan, though. I’ve been playing Spelunky a lot recently, and although that structure can’t be adopted wholesale, I think maybe there’s something there I can use. Will think about it, and write more soon.

Moreover, I’m away this weekend, and when I get back, I’ll be starting a new job at Slant Six Games. That doesn’t mean I’ll be unable to keep working on Wordsmith, just that it’s going to take longer than I had hoped.

Here’s what I’ve been working on today, anyway – the spoooooky graveyard. What objects would you make that can kill a ghost?

Keep the faith, folks.

http://zcint.co.uk/article/indie-week-wordsmith

No objects yet, but this landmass has about three times as much space as the demo map. For a sense of scale, look closely and you can just see “Wordsmith guy” in the middle there, next to the sun icon.

Oh, and I just got a day/night cycle going. (Bet you can’t guess what happens when you make a “sun”…)

Next up: scripted monster attacks after night falls! This thing is going to rock.

So… hi there! My name is Laurie Cheers. I’m running a one-man game studio that I call Cheers Games, and the first game I’m making here is Wordsmith.

I’ve been working on the Unity version of Wordsmith for about six weeks now. Here’s what I have on my “to do” list -

• Minor bug fixes – bridges and islands don’t position themselves over water correctly at the moment; letters can fall outside the available level area; when you kill the dragon, the game should end; you know, minor issues like that.
• More sound effects. For example, the “swoosh” sound works ok for the dart, but things like rifles, bombs and missiles are eerily silent at the moment.
• Better pixel alignment. Sprites are getting distorted because they’re not aligned to screen pixels correctly. Need to figure out how to fix that.
• Survival mode. The next step. So far, in the demo, combat has been fairly undemanding (with the obvious exception of the Dragon – who you can’t really defeat in combat at all). Survival mode will have a day-night cycle, in which you build up your resources in the daytime, and prepare for enemies to attack you at night. As long as you’re prepared, they won’t be too hard to kill… initially…
• And then Multiplayer? This is the long-term plan, only vaguely fleshed out. I feel like Wordsmith is crying out for some multiplayer options. Why does it have all those cosmetic items, if not to distinguish yourself from your friends and show them how creative you are?