I thought it would be a good idea to write down my experiences in my first game jam, identify what worked and what didn't, and what I'd do differently next time. There's a good chance this will be more useful to myself than anyone else but someone might find it helpful. To give you some context, I was tossing up with the idea of entering a game jam and the 7 day rogue like challenge was the next one in the calendar, so I had a read up on the requirements and thought I'd click the join button to see what other information was available - I was a bit surprised when the page congratulated me on joining the jam. I must say it was all a bit too easy to join, but much harder to complete!

That done I figured I might as well follow through and so set about coming up with a plan. Initially I looked at doing something classic, the warrior fighting an army of undead monsters through a vast dungeon but decided it was a bit too ordinary. In the end I went for a twist on this concept and decided to have a lone woman trying to escape from a  remote research centre on a distant moon. When you break it down it's pretty much the same thing with different models and textures.

So how did it go...

Well it started off well with most of the key visual elements built during the initial two days, dungeons were generating, doors were opening, the player could navigate around with mouse clicks, and I had created models for two enemies and the main character. From here things started to slow down as I had to balance work and family but I managed to get around 4 hours in each day of the week except for Friday where I put in 5 hours. Saturday morning was pretty mad, lots of last minute tidy ups, extra models to create, and bugs to fix, before submitting it right on my 1pm deadline.

This initial fast progress kept me feeling positive about completing the project however by Wednesday night I was starting to have some serious doubts with the realisation that there was only two short work days and one weekend morning left and so much still to do. It was at this point that I reviewed my to-do list and decided to cut a few items, the second enemy type had to go, the cut scene of the player escaping on rocket was out, and the much treasured idea of having the heroine combine hairspray with a lighter to create a DIY flame thrower was gone.

What went well...

• Reusing bits and bobs from previous projects helped a lot, there were times when I was tempted to rewrite things but fortunately decided against it. Being able to pull in a dungeon generator that I knew worked, a state engine that was quick and easy to implement, and other little code Jem's made life easier.
• Mind mapping the project and keeping a to-do list was useful to keep the focus, making sure I was doing things that were ticking items off the list rather than getting distracted on perfecting one little feature.
• Tracking my time went well, I kept a spreadsheet of every minute worked just like I would for work. I ended up doing a second working week with just under 37.5 hours worked on the project during the 7 days.
• Mixamo, I had never used Adobe's animation website before but I can highly recommend it, you simply upload your unrigged model, place a few markers for chin, arms, pelvis, and knees, and instant humanoid rig. Then you just select your animations from their large catalogue and download them each as separate .fbx files (without the skin). It did take a two attempts before I figured out it has a special .fbx format just for unity which helps with some of the quirks of unity and 3d animation.

What didn't go so well...

• Design/scope, the concept was ok but too hard to do well in the time frame, I ended up with a "finished" game that in all honesty doesn't feel fun. I feel it has most of the fundamentals to be considered a rogue like game however without the polish it's just a bit uninteresting to play. Different types of enemies, weapons, and inventory would have added a lot more depth.
• 3d, I knew from the beginning that it was going to be a big ask to do this in 3d but as my art skills are rubbish I figured I'd get a better result. From a visual point of view that arguably is true, however all of the other complexities that 3d brings to the table are real time syncs. For example I must have spent 4 hours alone on silly root motion animation issues.
• Sleep, working late nights and early mornings during the week didn't leave a lot of time for sleep.
• Random room content, this is one that still challenges me, it's easy to generate a random dungeon or maze but filling these random spaces with sensible looking content is hard. Some rooms look too cluttered, others too sparse, and after a while it all just looks like more of the same.

So what would I do differently next time...

• Do more design and research before the jam, identifying what resources I'll need and try to source them beforehand, rather than wasting valuable time trying to locate and/or create them during the jam itself. It's pretty frustrating spending an hour trying to find a free model only to give up and spend another hour making it yourself.
• I've got to get better at coming up with designs that are simple but fun, the less is more approach basically.
• Reduce the scope, I know everyone probably has this issue at some point but I think for me it's not about designing something that can be done, instead it's more about something that can be done and done well. When you're still fixing bugs 20 minutes before the cut-off without any time for final testing then you can't help but feel the quality isn't going to be up to standard.

The verdict

Overall I'd have to call it a success, I set an ambitious goal and achieved it, I delivered a reasonably complex game in a short period of time that is functionally complete... But next time I'll try something a little less ambitious - sleep is good :-)