TASVideos

Tool-assisted game movies
When human skills are just not enough

Submission #6604: DrD2k9 & ViGadeomes's C64 Decathlon "Max Score" in 09:57.0

Console: Commodore 64
Game name: Decathlon
Game version: unknown
ROM filename: Decathlon (1983)(Activision).crt
Branch: Max Score
Emulator: BizHawk 2.4.0
Movie length: 09:57.0
FrameCount: 35716
Re-record count: 2819
Author's real name:
Author's nickname: DrD2k9 & ViGadeomes
Submitter: DrD2k9
Submitted at: 2020-01-22 16:13:53
Text last edited at: 2020-01-22 20:26:34
Text last edited by: DrD2k9
Download: Download (3290 bytes)
Status: decision: cancelled
Submission instructions
Discuss this submission (also rating / voting)
List all submissions by this submitter
List pages on this site that refer to this submission
View submission text history
Back to the submission list
Author's comments and explanations:

Decathlon for Commodore 64

This is a score improvement to the currently published run.

There are two means of score improvement in this submission.

Though the total time for the run is longer due to more Pole Vault attempts and some minor delays for subpixel optimization, there were also some sections with time optimization improvements over the current publication.

  • No encode this time...if someone feels it's absolutely needed, I can add one later

Hitbox Management

ViGadeomes contacted me because the pole-vault situation in my previous submission was bothering him. He had a theory that it was indeed improvable (it was). Essentially, improvement was accomplished by altering running speed and location of the pole plant to better manage the location of the hitbox changes as the player goes over the bar.

ViGadeomes initially thought this had something to do with acceleration at the end of the run, which led me to think about something else I hadn't considered the first time around...

SUBPIXELS!!!!!!!

Never forget the importance of subpixel position!

By modifying the run portion of the various throwing/long jump events I was able to delay the throw/jump until the last possible sub-pixel before a scratch would occur. This further improved event performance and final scoring.

Performance/Scoring Changes By Event

100m
  • No Change
Long Jump
  • Old Distance = 9.37m
    • Old Score = 1284
  • New Distance = 9.55m
    • New Score = 1319
Shot Put
  • Old Distance = 23.57m
    • Old Score = 1260
  • New Distance = 23.59m
    • New Score = 1261
High Jump
  • No Change
400m
  • No Change
110 Hurdles
  • No Change
Discus (Distance improvement, but no accompanying score improvement)
  • Old Distance = 71.55m
    • Old Score = 1239
  • New Distance = 71.56m
    • New Score = 1239
Pole Vault (The new score now crests 1000 so we get the trumpet fanfare)
  • Old Height = 4.6m
    • Old Score = 957
  • New Height = 5.0m
    • New Score = 1052
Javelin
  • Old Distance = 100.96m
    • Old Score = 1210
  • New Distance = 101.5m
    • New Score = 1212
1500m
  • No Change

Total Score

  • Old = 12,094
  • New = 12,227

Timing Optimizations

ViGadeomes will explain

Final Thoughts

This submission is a good example how even a previously well-researched and perceived Max Score run can still have potential improvements. This mimics the situation we frequently see with runs focused on speed instead of score. What we may previously believe to be the fastest possible, is often improved upon. Similarly, what we had previously believed to be the best possible score was indeed improvable.


DrD2k9: Cancelling. We realized that some of the High Jump inputs were sub-optimal for time. It appears this may be due to the C64 dropping inputs. Thus further timing optimization may be possible in other events as well.


Similar submissions (by title and categories where applicable):