A follow up to my post on Autonomy, Mastery and Purpose on being the keys to a great game.

The experience of playing the Duke Nukem Forever demo made me ponder the idea that Autonomy is more or less dead in the first person shooter genre. Every modern FPS I have played (with the possible exception of Crysis) seems to have replaced the idea of leaving the player to solve the game their own way with the idea of holding their hand all of the way through it.

Let's look at the ways it has done this:

Level design

(Note: I have no idea who originally authored this map. It's pretty good though. I think the map is one of the later levels from Doom 1's shareware episode.)

Back in the "good old days", the player was simply dropped into a map that resembled a place - a maze-like Castle floor in Wolfenstein, a Sci-Fi or Fantasy environment in Doom, or a city block in Duke Nukem 3D - and left to explore it their own way.

Granted, this wasn't always fantastic. Exploring mazes became monotomous, especally when backpeddling and finding keys was involved. But at least levels with multiple paths give replay value. The first level of Duke 3D springs to mind, although it isn't the most open level in the game you could siege the movie theater from either side, and there were numerous optional side rooms that could be explored.

Recharging health

In most modern First Person Shooters, you can be fully healed after taking damage by just stepping out of the way of gunfire for a few seconds.

I admit that it was sometimes frustrating to hunt down health kits. But the flip side of it was an intense experience, where you had to be extremely careful with your health, especially as it was running out. It was also very satisfying to find health after holding on at 10% for awhile.

I'll be fair and say that I actually like the concept of recharging health in multiplayer modes - to a degree. The first time I played Goldeneye multiplayer, I couldn't score a single kill as the game's owner knew exactly where the armour was. Recharging health removes the requirement for players to learn where the health kits are in order to play competitively. By the same token though, finding and jealously protecting health kits did add another layer of skill and strategy.

Restricting the number of weapons

Most modern first person shooters only allow you to carry two guns at a time. By contrast, all of the classic ones allowed you to carry every gun in the game simutaneously.

Sure, this was far from realistic, but it was great to use any gun in the game at any time. Asides from being fantastic for replay value, it added a layer of strategy as often you'd have to weigh up using the BFG-9000 ammo to get out of your current situation, or saving it for a more heated battle later.

Quick time events

I don't think I need to elaborate on this point. No body seems to enjoy these.

The end result of all of this is:

When I get a first person shooter these days, I play through the single player mode in around eight hours. I have a great time, but that's the game over and done with, because there's almost no replay value. Eight hours doesn't even come close to the amount of hours I've spent playing and replaying Duke 3D, Shadow Warrior, Doom, Goldeneye, Blood, ROTT et al. I think even the free Shareware versions of these games had comparable play value to their modern counterparts.

Long story short, I failed. But I found something that works better.

About a week ago I started hunting for a simple issue tracking system that runs on the LAMP stack. The main reason was to help manage the enhancements and bugs for my various gamedev and web projects. These were my requirements:

• That it is simple and lightweight. Usuability, especially when issuing and managing lots of tickets is paramount.
• That it is slick. I used to use Drupal derived Open Atrium, but I found it took too long switching between pages to bother with.
• That it is multiuser. The ability to restrict access to projects (especially to the general public) is an absolute necessity.
• And most importantly, it runs on my cheap, shared hosting Linux environment.

I announced my intentions on Twitter, and immediately got a response from a friend that a simple issue tracker doesn't exist for any stack.

And I found that he was largely correct. Although I like BugTracker.Net a lot and that would have been perfect, the '.Net' part made it unusable. Various others I looked at, though they ran on the LAMP stack were unsuitable for shared environments. I almost settled on FlySpray or Mantis, but found FlySpray's administration was far too clumsy and Mantis was far too overloaded with data.

There were some commercial hosted solutions, like 16 bugs that might have been fine, except that I didn't want to have to pay a reoccuring cost. I find it somewhat ironic that some commercial software products get their value by having less features than the open source equivilent.

Dwelling on it further, I think part of the problem is 'simple issue tracker' is an oxymoron. Instead of looking for a simple issue tracker, I should have been looking for an advanced task manager. TaskFreak! was what I settled on. It's dead easy to define projects, define tasks, assign those tasks to people, add additional info and mark them as complete when they're done. It's also very, very smooth, thanks in part to AJAXy goodness.

Granted, TaskFreak is a bit clumsy in some places and is missing some stuff I'd really like - such as Email notifications, Milestones, and File Uploads - but it more than fits the bill for my requirements.

New version of WitchBlaster is out, now with two player mode. Play it from the WitchBlaster page. (Note there's no downloadable version yet, need to resolve some issues).

I've been tinkering with a few of my game projects for the Earok 2011 collection.

WitchBlaster

Awesome new artwork from Anthony. He's replaced my Bat with a brilliantly cute one, in three variations (Regular, Kamakazi and Hip Hop - Not depicted in the screenshot). Kirsty noted that the new Bat was too cute to shoot at. Also a palette swapped version of the Witch for player two.

Two player mode is coming along nicely, except there's lots of minor glitches here and there that need to be ironed out. Definitely very playable though.  It's kind of fun to try controlling both characters at the same time.

Heart of Ice

This is not an actual screenshot, rather a conceptual mockup of the Monkey port. In the end I decided to go for pseudo-isometric instead of faking third person 3D. Also there will be dialogue between the two characters throughout the game. The actual game itself won't change much, except there's definitely going to be a two player mode, and I may make the game much more frentic. I just wish I could get some better artwork, though!

Noah More Heroes

A few bugs have been fixed, plus a custom build for the Winnitron NZ has been produced and sent out. No word yet on whether or not it actually works.

Firestart

I got an email the other week asking for a two player mode, so it's definitely going in. Hard part is working out the ideal mechanics, even just doing a straight two player co-op mode could be tricky as it was never designed to handle more than one player character, and the scripted events expect the player to be in one place. Also, there's a new story for the game in progress.