jam postmortem

no design thoughts here, just implementation/project details

• this was an exercise in rebuilding my own confidence more than anything -- did it work? mmm hard to say
• my approach to the 10-day game jam was to establish a very strict timeplan and hold to it -- day 1 for player systems, day 2 for stage enemies, etc etc. while exhausting, i was able to do a lot in the time allocated, though at times i was certainly "hanging on by a thread"...!
• this was also an exercise in using 3d modelling for stage/boss assets -- i've been learning blender for about a month now, and this project was a baptism by fire. i think most of the stage assets are very ugly, but they are there, and can be improved upon. i also didn't have much of a clear vision for what i wanted in the stages, so they just look a bit like mush -- though i'm proud of the st2 boss background
• otoh, i was extremely pleased with the boss assets i managed to make. ENDEAVOUR is a little simplistic but could easily be expanded with destructible armour plating etc, the DELEGATES (st2 midboss) and THE ORB were fairly simple creations using free meshes as a baseline, but i think they have lots of character and look very nice
• on the programming side... i do write a lot of code. there are 7400 lines in this project not including the base framework stuff, libraries, etc, just game logic. i think i'm fine with this though, i don't mind repeating myself if it makes it easier to tweak very specific behaviour later
• custom collision system based on a bunch of spatial hashes -- conventional collision systems piss me off when you have to say run a function on every two objects to tell if they should collide or not. i'd rather just say quite obliquely "okay check player bullets against enemies, now enemy bullets against player, etc". it seems to be good enough
• i want to move my style more to using named arguments (ie the function{} syntax, everything just goes in a table in typical lua fashion) -- bullet pattern system uses this style and it was pleasant to work with -- is there a disadvantage?
• this project had a target framerate of 144hz. i don't think a high refresh rate does a tremendous amount for STGs, and better suits stuff with a very dynamically moving camera, but it does make a marked difference when projectiles get fast, and it improves responsiveness
• i inadvisedly thought i could try to do my own music, but music is just far too important in STG, so i borrowed some of my favourite modules from modarchive again. most of these modules were written in the 90s or so -- as far as i'm concerned, it's a totally free game and all use is attributed, and if someone wants their music gone -- it's gone. i think that's fair. i don't really want to go trying to track down the original composers to ask explicitly for permission -- but in any case, if any of them read this, thank you for your work
• sfx were left until the very last minute again and it probably shows. another thing i would like to take a more active role in, but at the same time, can't do everything, and i don't think audio of any type is really my skillset