Shortest Postmortem Ever

[davidyu]

Basically, everything went well from the get-go.

[davidyu]

But seriously:

The good:

• We picked a very simple mechanic and a robust engine that supported our genre (platformer) right out of the box.
• We got to our first prototype within two hours of starting.
• We spent more time designing levels and tweaking them than implementing new features and mechanics (double-edged sword)
• We didn't work long hours (average per person: ~3 hours on Friday, ~6 hours on Saturday, ~4 hours on Sunday)
• We pair-programmed!

Room for improvement

• SFX design. They were an afterthought and could use more work.
• Music, which did not exist in our game
• We could have spent more time tweaking the core mechanics (jumping, moving). A couple of people specifically felt that movement and physics were too floaty.
• Better workflow when editing Unity levels. We stomped over each others' work several times. This just comes with practice.
• We could have thought of a name before we submitted :smiley:

Nathan, add anything you think I left out.

[degrand1]

The merge conflicts will definitely go a lot smoother in the future now that we have a better idea of what files get changed whenever values get tweaked in Unity. Merging scripts didn't seem to be too much of a problem, so probably the best approach is to pair program until we get far enough to begin work on separate levels.

I honestly had a lot of trouble creating basic assets on Photoshop as well. I think it'll be worth it for me to just start iterating through random Photoshop tutorials until I get a better idea of how to use the program.

[davidyu]

Yeah, art is a fun problem. I take back my original thought that having an artist to shoulder the burden is always good in jams like this. Because we didn't have an artist, we could do whatever the hell we wanted with the art assets. And the weird faces came about directly as a result of having that freedom ("hey, let's mess with the players a bit. They don't expect the protagonist to flip over inexplicably. They don't expect a box to be expressive.")

Stepping back, I think what we have here is a great prototype for something (slightly) bigger. If we wanted, we can easily go deeper with mechanics:

• as we originally planned, make each cointype bestow you with a special ability that allows you to finish the level in a specific way
• more hazard behaviors: moving incompatible coins, AI-driven incompatible coins

I feel that each of the above would open up many doors for fresh new levels. If we target 10-15 minutes of gameplay in total, we can have a reasonably fresh game that should definitely exist, and it would be one that I feel many people will enjoy.

Let's take a few weeks off first, before jumping any guns.