Do you smell roses? What went right:

For Travis:
Building Fools Gold had us throwing ideas against the wall and seeing what stuck. The process was fun and delightfully naive, and we have a number of good laughs about it. The tools were touchy, and the code was foreign but I did manage to get some really killer AI working with a lot of time allotted for looking stuff up in the documentation and searching the forums. I also got my hands pretty dirty with the C++ code, but ended up ditching it because I didn't want to figure out compiling on the Mac.

Adding sounds turned out to be really fun, despite the total lack of proper equipment. Everything was basically done by speaking into my USB headset microphone, and subtracting the ambient noise from a "silence" sample, or stealing wav's off the net. And so it ended up that all the spells have movie or song quotes of some sort ("I'm Invisible" from Cowgirl by Underground made famous in the movie Hackers, "I know Kung-Fu" from Neo in The Matrix, "Run Away!" from Monty Python's Quest for the Holy Grail, and "Booby Traps" from Data in Goonies), and all the sound effects were spoken words of some sort. I laughed up a storm every time I put a new sound in the game and make it trigger off a button press.

The game ended up being pretty empty, but you could envision a lot more going into it... not that it would have made it better. It's giggly funny for about 15 minutes, but nothing will hide the fact that it's a turd of a game with no real reward for playing other than to dodge Blizzards. Overall Torque was extremely impressive, and when we are able to properly own the code it will be an awesome tool. But it's got some serious limitation that you have to really understand before you get too deep and find yourself up to your eyeballs and no way to get back to shore...

For Marcus:
Once you get a handle on the world building tools, Torque is an amazingly easy thing to use. Tools like shape and foliage replicators were a godsend. Once I got in there and started building and painting the terrain it was as easy as drawing. Adding content? Pretty much a breeze. Certainly some quirks, and I am are still trying to come to grips with the why’s of certain things in Torque, but over all it was great stuff. The particle system, once you wrap your head around it, is fantastic.

Pulling together any sort of playable game, or toy, in a week was an accomplishment. I am strangely proud of this game that is good for minutes of pure awesome, but not much more. It is the sort of activity that would be fun to do on a regular basis, especially in a long large production cycle. Taking seven days off to get some quick rewards and progress was nice. Plus, it makes fantastic smalltalk…

For Sean:
What seemed most instructive here is looking at how we worked together as a team. I am both encouraged by what I saw and feel that much work needs to be done in that regard. I think our three skill-sets do have a very nice overlap and our knowledge pools are generally complimentary. It was fun to see these stupid things come together in what turned out to be much more impressive then I expected. The time spent on the landscape makes everything else feel much more complete then it is, and hey, this thing is actually playable!

Plus, it really is hard to avoid polishing the turd! - I still want to go back and tweak some things...

What a stench! What Went Wrong:

For Travis:
The spells were where I really got bit because the architecture of the scripting language was still a bit foreign despite the similarities to existing languages, and held some hidden gotchas. Blizzards also proved to be a bit more challenging than we had hoped, but we found a way around it. I'm not totally convinced it's our unfamiliarity with the code which drove us to such extreme hackery either... as best I can tell it's one of those things where if you want to do things like mount particle emitters to moving objects you gotta play by the rules, or change the engine yourself, and that's the accepted solution. But maybe I'm way off, and the answer is really elegant and we just don't know it.

Day 5 and 6 were a blur of coding and searching forums. I spent about 4 times as much time trawling for how to implement solutions than I did coming up with them as I struggled to overcome my learning curve. During this time I made some important discoveries that I'm still digesting the implications of... and waiting for the other shoe of my understanding to drop, since I can tell I'm still missing something.

For Marcus:
Lets take a stab at the things that went wrong for me, thankfully there were not a lot of them. First and foremost for me, as usual, our development tools suffered incomplete documentation. This is always something that drives me nuts for any product that has been around longer than five years. At this stage of the game someone should have sat down and done a detailed documentation on the product and tools. No excuses for sections of the software documents to sport lines like ‘This Needs To Be Filled In Later’. ( Yes, you will find those very words scattered throughout the Torque documentation and development site. )

Struggling through esoteric tools and dredging forums and Goggle for solutions got me through most of the worst problems. Mucking about and clicking around got me through the rest. It did chew up a vast majority of my time for the first three days or so. The Torque GUI editor also gave me heart palpitations, especially since we tackled UI at a very late stage in the development cycle, not a mistake one should normally make.

Being a creative person digging around for visually appealing content was important. Finding something to bring things to life, something to actually –look- at so that we all felt some sort of progress. It took me a couple days to fully understand how ugly our game was going to have to be. Seven days is not enough time to make a polished gem, and the running joke on this project quickly became the idea of ‘polishing the turd’.

My mantra: we have only seven days… we have only seven days…

For Sean:
Communication problems caused issues throughout the process such that though this is a 7day throwaway much stress was generated. Much of this can I think be solved with a better work flow, better (or any) pre-planing, knowing how the tools will slot together, and better pre-planed division of labour.

Tools in general can very much get in the way. EXPECT to have at least one person "wasting" time on keeping the tools running and solving issues. Torque is both very impressive and highly frustrating. - again, choose your tools carefully or go in knowing what to expect. Most problems did indeed get solved faster with three different mindsets then with one mind and google.

Also, the end goal needs to not be a moving target. Having a floating target of how many features are going to exist is fun but makes it hard to set priorities so that things get done on time. Plan the work out ahead, then do it.

Please make it end! Conclusion:

Torque Game Engine is a great tool for rapid prototyping and getting something put together in a hurry. It has it's quirks that you have to struggle with, but that seems to be pretty standard with any tool set. The really important part of this whole project was to make a game in seven days, and we certainly succeeded in that. It is a really bad game, but it is a game. It also makes a dandy joke...