My published runs.

Tom and Jerry was my first published TAS. Tom and Jerry was the 3rd game I owned for Gameboy, and the other 2 - Pokémon Blue and Super Mario Bros. - already had runs. So I made the run of Tom and Jerry. The first version was not very optimal - the time would probably be beaten unassisted if TAJ speedrunning was as popular as Super Mario Bros. running. The present movie could be improved by a couple seconds by faster jumping, better luck-manipulation in level 4, and not landing above the exit sign in level 9 (at about 8:13 in the published movie). However, I've had a lot of trouble with the RNG in level 4, enough to negate all other improvements in my attempt at obsoletion.

River City Ransom is one of my all-time favourite games, my favourite NES game, and the game I consider most fun to TAS. Because of the diverse fighting style, optimizing a run of RCR is very hard. My present run could probably be improved up to 1 minute (this on top of a 2-minute improvement on the first run) by better boss fighting. However, because different styles have different results boss-to-boss, a "wiki-style" run would be best. That is, a run where several people try to optimize each boss fight and the best result is taken.

Subs-1066up

I have several rejected submissions. The first Tom and Jerry run was unoptimal, and was beaten. It has nothing notable, other than that it is my first TAS and was made in 3 days (Christmas Eve through Boxing day).

However, the Jurassic Park III: Island Attack run does merit some discussion. JP3 is another game I owned which I thought would make a good run, and was the second game I TASed. I was very careful with optimizing turn timing and button pressing, and lag notwithstanding, it is frame-perfect. That said, it could likely be improved several seconds, maybe even 30, with better understanding of lag. However, this would not help much, as it has 2 auto-scroll levels and 5 unavoidably long boss-fights. I should look for a glitch similar to the pause-glitch in Megaman. If I could do extra damage to bosses, or get a one-hit kill, the run's time would be nearly halved and I believe the run would be acceptable.

Serpent was rejected due to bad game choice. It was a Tron-like game, and I simply rushed into fight mode and won in seconds. At 7 seconds, it is the fastest TAS on the site (as far as I know), and it was intended to obsolete the 10-second King's bounty run, as a TAS which was fast for the sake of being fast, the specific game notwithstanding.

Baseball Simulator 1.000 was the only run I made which was entertainment-geared. It simply showed off several strange abilities in the game. It is my favourite rejected submission. I think as an entertainment-based run it is good, but it does not meet with the goals of tasvideos, as it does not win the game. By scoring 1 more run I could finish the game by the "mercy rule", and then I think it would fit into a "pure entertainment" category. Which brings me to my last submission.

River City Ransom in 6:40. This run threw entertainment to the wind for an input file 13 seconds shorter. This was one and a half minutes faster than the best RCR TAS, and over 2 minutes faster than the published movie. However, it had one problem: The final boss, rather than simply being beaten up, committed suicide. Seconds after entering the arena for the last fight, Alex throws Ryan through a wall. Input end. Simon, the last boss, beats up Alex, then turns on Ryan. Lacking great knowledgeability of cause and effect, or basic pattern recognition skills, he tries to get at Ryan by attacking the wall. He throws various objects at the wall, and is surprised when they bounce off his head and kill him. It takes about 3 minutes, as he often gets stuck in "loops" where he does nothing for a while.

There was another RCR submission, which was the polar opposite of my 6:40. #1740: adelikat, JXQ's NES River City Ransom "Playaround" in 12:55.38

This run spent a long time goofing off, playing baseball, stalking girls, and bouncing off walls. It sparked a fairly large debate about speed vs. entertainment, and the creation of a playaround category, especially when it was suggested that it obsolete my movie. I believe no slower movie should ever obsolete a fast movie. However, I would support an entertainment-only category for movies like this and my Baseball Simulator run.

I would like to see a lot of runs, but here are the major ones:

Whenever someone says something about how fast a TAS can or can't be, I mark it here.

Glitcher: Mark my words, next year the bar will be lowered to 4:00 - 12/14/07, on Super Mario Bros. (This prediction was incorrect, but the runs have been made a little faster several times since that proclamation)

andrewg: This run will never be improved again, you can quote me on that. If you don't know me, let's just say I know more than the average person about this game ;) - 1/1/11, Super Mario Bros.

Mr. Kelly R. Flewin: I predict a final time of 4:56 myself. - 11/22/04, Super Mario Bros.