It was a fun watch, and enjoyable to work on a gamejam for such a limited platform. It's always impressive to see what others are able to squeeze out of this platform.
As far as Fencing goes: the reason I didn't finish collision wasn't due to the animation system. Early builds (before realizing my other issues) had the sword only one DXYN wide, so erasing and redrawing just the swords in separate calls wouldn't have been an issue and I could just check the collision flag.
The problem is that I wanted to handle sword->sword collisions differently than sword->player. And while I could probably do a dance of erasing both swords, stamping each one in for collisions, and then stamping in both, it seemed like it'd be more efficient to do things by hand. I was sensitive to Display Wait since VIP-playable was a stretch goal of mine.
So if I wanted to do everything by hand, I needed to know what the valid bounds were. And of course my sprites usually don't reach each other, so it didn't seem worth it to put in that effort until the sprites are fixed.
I'm probably going to have to get over my pride of having it all be my own work and just ask someone I know with artistic skill to help out with the drawing part. Then I can tackle the rest of it.