Plan for Jay Barnson

A Game In A Week - HACKENSLASH, the RPG!

One Man. One Week. No Engine. No Budget. IT CAN BE DONE.

I still don't have the full "diary" ready yet - but here's the quick-and-dirty on Hackenslash - the latest (but definitely not greatest) product from Rampant Games. Available Absolutely Free, and worth every penny! You can download it and ... well, hopefully enjoy it, though I'm making no promises, at:
www.rampantgames.com/hackenslash.html



As I recounted in a previous blog, Tom Bampton challenged me in a thread to put my money where my mouth was. I decided that trying to make a game in only ONE WEEK (which I defined as 40 hours of dev time - a traditional work-week), starting from square one with no pre-existing engine, and no budget (and no pre-owned software tools that actually, you know, cost money) wasn't hard enough.

So I decided to fly in the face of conventional wisdom that indies can't do RPGs (Josh Ritter's fascinating project, and Spiderweb Software's incredibly successful games notwithstanding), and make in an RPG. Arguably one of the most complex kinds of games you can make.

Of course, I went for bare-bones ultra-simplified 2D gameplay. In spite of that, it was still pretty dang complex. In the end, there are several features that I *coulda, shoulda* done if I only had a few more hours. There's no sound, and my brother even whipped out some music tracks for use with the game. I have a vestigal quest-system in there that never got used (instead it's pretty much a level-treadmill game... see how far you can go before you die). I had some really nice complex logic for room features that were only used to a fraction of their potential. There's only one monster, and it would have been EASY (but a little time-consuming) to throw together a dozen or so. Some of the skills didn't end up getting used (I had to get rid of the merchant at the end for lack of time, so the Haggle skill is useless). There's a whole structure in place for doors separating portals that isn't used. And secret doors and hidden items - that would take only about 30 more minutes to complete... most of the logic is already in there. And the monster movement is buggy. And there are only six rooms in the dungeon (give me an hour and I could triple this number! With cool features!) And it would only take about 30 more minutes to finish the monster spellcasting. And we won't begin to talk about the complete lack of play-balancing. Or testing.

And...

And...

And...


But it's a game. And it's an entertaining diversion for a few minutes. And there's a lot of things that went into it that I'm proud of. For example:

* You can bribe monsters - sometimes it's your best defense (especially against multiple opponents). I really wanted to add some kind of non-combat element to the game. Of course, if your Negotiate skill sucks or you offer too little money, they'll get offended.

* I added random treasure generation - so you can have all kinds of different kinds of equipment with different magical effects. Mix and Match!

* I ended up with 5 spells at 4 power-levels each. I hate games where spellcasting is weak and very limited. Unfortunately, it's probably a little overpowered in this game.

* The entire game was written in Python, a free language. I used PyGame for graphics and input - which is in itself a Python binding for SDL. I used "py2exe" (and a lot of swear words) to compile the game into this distribution, so that you don't need to install Python to play it. And of course, all these tools are FREE and available over the Internet.

* Graphics were done in The GiMP, an open-source graphics package.

* Some of the graphics (particularly floors and walls) were obtained from free-texture sites like Toob's Tiled Textures and Mayang's Free Textures. In most cases I had to do a lot of editing in GiMP to make them work for the game, which took a surprising amount of time.

* The game includes some other RPG conventions like trapped chests, leveling up, resting to regain health (only available in the starting room up the stairs).

* I actually spilled over 40 hours by a few minutes. And I didn't include time I spent documenting what I was doing, nor did I include time spent packaging up this game for distribution (which was a learning experience by itself) or fixing some of the crash bugs I found while getting it ready to distribute.

Some lessons learned:
#1 - I always underestimate the amount of time it takes to create art assets. Maybe it's because I suck and I'm not content with the type of stick-figure art I actually COULD do quickly. If I had an artist helping out, I could have saved myself a good 6 hours or so for finishing up the tons of features that are only half-implemented

#2 - I imagined an XML-style, data-driven means of generating game content. That didn't happen. If I was doing it over... well, if I was doing it over and didn't HAVE the 40 hour time limit, I'd spend a lot more time on that working on at least a rudimentary editor.

#3 - I ended up completely scrapping some of the interfaces because I realized they were useless and annoying. For example, I used to have two menus to pick up an object - one to attempt to pick it up, and then another to ask you if you wanted to replace your currently equipped item with this one, or simply 'cash it in'. That's an extra level of clicking that's just annoying - better to put all the options up-front and keep the menus as shallow as possible. Put the brains in the UI where possible, rather than forcing the player to tell the dumb computer what to do.

#4 - It takes longer and longer to code each new thing than the thing before it. Yeah, I already knew this rule. But simply making sure a new system works well with everything else that came before it is a chore that really slows down development on a larger game.

#5 - I waste a hell of a lot of time when I'm developing. I was very scrupulous about only timing myself when I was actually coding - any break I took more than about 10 minutes was not counted. So as a result I had an almost IDEAL 40 hours in which to do this. But I discovered how little "real" work I get done in an evening, with time spent surfing the 'Net, answering email, playing a quick round of Unreal Tournament 2004 or Orbz, and whatnot. I was aghast one night when I realized I had only gotten about 2 hours of work done in over four hours. Of course, some of that time was spent documenting, which was important (more on that later), but certainly not the bulk of it.

#6 - Python rules

#7 - Murphy's Law will conspire against you to totally kill productivity and opportunity on the one weekend you intended to devote a great deal of time to a project.

#8 - A feature-full engine is NOT ESSENTIAL for game development. It can be done from scratch - it just takes effort and no-how. However, a solid engine like Torque2D would have REALLY helped speed development, I think.

#9 - 40 hours isn't that much time. Really... a game-in-a-day is only 24 (assuming full productivity), so 40 hours for one guy is about like 2 guys doing a GID. Of course, I coulda done the Electronic Arts thing and just expanded the work-week up to 60 hours (or more)... but dang, what would have been the point?

#10 - Being an indie, and having the freedom to just do something like this on a whim "because it's cool" just plain ROCKS!!!!

Anyway - if you are interested in trying out this game --- well, don't expect much. But the link is here:
www.rampantgames.com/hackenslash.html

Enjoy. And let's see more people whip out a game in 40 hours.

Jay Barnson
Rampant Games