Having recently gotten a little bit of spare time, I've taken a bit more critical look at progress so far. It seems that I find something that can be improved every time I get 2 or 3 stages into the game. Also, "josh l."s run has provided some ideas that were readically different than any of my ideas. His WIP can be found at
http://omake.studio-zoe.com/Kickle%20Cubicle%20(U).fcm and is posted in the "Top games that haven't been done yet" thread, post number 31.
(0) I plan to continue with Kickle Cubicle instead of Meikyuu Jima. The transition between levels is a bit shorter, but all of the levels have been changed. The levels are just too different. The levels that "are the same" have often been moved to different locations and changed with more and/or different enemies. The story and number of levels you must complete to advance the story is unchanged, however. The bosses take more hits to destroy. Overall, I think the two games are too different to justify doing a Meikyuu Jima run instead of Kickle Cubicle. It's a shame, really, since the lotus roots used at the end of a couple of stages in Garden Land look pretty cool.
(1) Opening menu can be done around 400 frames faster. I can chalk this one up purely to inexperience.
(2) My method of walking diagonally whenever possible is still the fastest means, but a faster way to go directly in a cardinal direction is to go 6 or 7 frames in that direction, then 1 frame perpendicularly, then 6 or 7 frames in that direction, and so forth. I refer to this is "whiplash walking". This discovery meant an immediate start-over on the run without hesitation.
(3) There is a variable amount of frames between levels, and it can be manipulated. I consider five parts of the "between level" sequence, recognized by whether the screen in blank or not:
-A- The frame when the screen first "shakes" upward and you have lost control of Kickle up to but not including the frame that the screen finished going black. I have seen this pary vary from 583 to 600 frames in length.
This length appears to be primarily level-dependent, but a couple of levels in my posted WIP and "josh l."s WIP differed by as much as four frames. My only conjecture thus far is that the number of enemy creatures on the screen affects this length in some way.
-B- The blank screen time between the level and the map. The length of this time seems to be 65 to 69 frames, and the only way I've found to manipulate this length is by varying the controller input on the last frame of controlling Kickle prior to the part -A-. This frame of input is otherwise inconsequential as far as I can tell... (knows something he isn't telling yet)
-C- The time on the map. 154 frames. This time appears fixed and consistant.
-D- The blank screen time between the map and the next level. 85 frames. This time appears fixed and consistant.
-E- The time with Kickle and the balloon before you gain control of Kickle. I still need to do more testing, but the time is around 350 frames and seems to be affected by Kickle's starting location.
Wow... that's a lot of consideration for 8 frames over a span of nearly 21 seconds. However, (8 frames per level) * (100 levels) = 800 frames = 13 1/3 seconds.
(4) I've finally understood the cause of some control issues similar to lag. It seems that the game fails to recognize any controller input for 2 frames when an enemy is "hatched". Instead, the game reacts as though you had held the input for both the frame of input and the following two frames. I speculate that this occurance is due to the change from a static sprite of the "enemy egg" going through frames appearing to "hatch" an enemy to the static sprite which is overlaid by the actual enemy. Thus, the game fails to recognize controller input for the first two frames that the object enemy first comes into existance on the screen. This is just a pain that seems unavoidable.
(5) I'm still working on route optimizations, but some of my routes are rather sub-optimal when the "whiplash" walking technique is used. I'm always up for suggestions, advice, and demonstrations!
Sorry for the long post, especially without including a WIP, but this run hasn't been forgotten. I will hopefully have a good WIP in the next week or so!