Submission #1069: CtrlAltDestroy's NES Boulder Dash in 19:42.32

Nintendo Entertainment System
baseline
FCEU 0.98.15
70939
60
7608
Unknown
Boulder Dash (U).nes
Submitted by CtrlAltDestroy on 5/1/2006 6:38:43 AM
Submission Comments
After several months of being stuck on Volcano-3, I randomly tried again tonight and somehow got the amebas to convert a whole 2 seconds earlier than last time! Lucky!!
Movie Info
  • Created with both FCEU 0.98.12 and 0.98.15
  • Aims for fastest time
  • Manipulates Luck
  • No deaths
  • Abuses programming erros
  • Plays on hardest difficulty (Password 724045)
Fun Facts about this movie
  • This run is exactly 1523 frames ahead of the last run, for a total improvement of 4696 frames over the first run.
  • All the levels in this run but the last two were completed within the course of about 2 weeks, but Volcano-3 took almost a year to finally finish.
  • The time improvement came from strategy improvements on almost all of the levels- most small, some huge. The four improvements over 100 frames were Sand-1, Volcano-2, Volcano-3, and Volcano-4.
  • Sand-1 is 3 seconds slower than in the previous run according to the game timer, but 3 seconds faster in real time due to lag reduction.
  • Volcano-2 uses an interesting new trick where I physically manipulate the scorpion into place instead of waiting for it.
  • Volcano-3 (The Ameba level) is slightly more interesting to watch than last time (but just slightly!) Let go of fastforward when the timer hits about 158 to see it.
  • Volcano-4 takes advantage of a new trick. Due to a programming error, squashing a bug will cause the rest of the bugs in the stage to freeze for a few turns. I use this to drop a rock on a bug that otherwise wouldn't have hit it.
  • Ocean-2 is the only level in this run that is slower than in the previous run- by 16 frames (1 step). This is due to a luck manipulation error I didn't realize existed- The first batch of diamonds refused to become collectable for 32 frames, whereas previously they became instantly collectable. Nothing I tried seemed to change this. I did, however, manage to improve this level's execution by 16 frames, so as to be behind 16 and not 32 frames.
  • The previous run was recorded from reset instead of power-on. When the run was published, Bisqwit (or one of his scripts) changed the movie to start from power-on, causing the currently published fcm to desynch at Sand-1 due to lag. This new movie correctly starts from power-on.
  • My completion time is the name of the prequel to the NES game I last submitted a movie of.

Bisqwit: Works without desync, seems quite difficult, and it is an improvement to existing movie. Accepting. That guy must have infinite pockets. Storing thousands of diamonds the size of himself. That second-last stage is weird btw, with that long wait.
Last Edited by on 1/1/2022 6:13 PM
Page History Latest diff List referrers