• Aims for fastest time
  • Abuses programming errors
  • Manipulates luck
  • Takes damage to save time
This is a run that attempts to beat Mario 64 in the fastest possible time. In doing so it beats the previous movie by 55 seconds. Our summary of how this was done follows...

Swordless Link's comments...

The original goal of this run was to achieve a time lower than 6:40, and thanks to several timesavers which AKA mentioned to me (the glitch we used to access Bowser in the Dark World, the glitch we used to get from there to the basement), this seemed more than possible. I also had some ideas of my own, which we put to good use, such as the alternate route in BitDW which saved about a second over the older route. The obvious thing to mention about this run, however, is that it finishes the game with zero stars, something which was previously impossible. It all started out with a theory which AKA had, which involved performing a backwards long jump on the stairs in the basement, and getting Mario wedged inside one of the pillars in the room with the Dire, Dire Docks painting, just like what we did after exiting BitDW. This proved impossible, as Mario had way too much speed to be able to move to the sides of the room. It was at this time that I began thinking about adding an extra long jump outside the door, because logically, this would still cause his speed to increase, and then did another one when I was inside, as I realised that this would cause his speed to increase even more. This gave me enough speed to break the DDD painting, but there was only literally one frame I could've long jumped on while inside the DDD room to have enough speed to break it. This discovery of mine took about 45 seconds off the overall time. Overall, I'm very happy with this run, it definitely seems a lot more optimal than the other runs which AKA and I have co-authored, probably because we re-did so many things so many times in this run.

Mitjitsu's comments...

During Rikku's submission, I wasn't too interested in doing another run. I decided that I'd only work on another run if there was a way to notably improve the current run. Firstly, I was utterly convinced that there was an alternate lobby glitch which was faster. It took countless revisions, but I eventually found one that worked. Secondly, I thought there would be a way to get quickly to the basement without re-entering BitDW. And last of all, I had a good theory on how to bypass DDD, although Swordless Link was the one who eventually found a way to do it, and it was a lot faster and smoother than I ever could have been conceived. Also, I knew that a few frames could be saved if diving was used over long jumping where applicable, although that alone wasn't enough to warrant another run. The main timesaver that I found was how to get to the basement door quicker. Although, again, I would never have thought that it would have turned out that good. We put an extra special effort into this run, but it was worth it in the end, and I hope that most of the viewers will agree. If you calculate the number of re-records per frame, then it has one of the most dense re-record rates on the site if you base the re-record rate on re-records per frame (since the game is 30fps), then this run would be third overall if you discard the ALttP glitched run.

How it came about...

AKA (21:34:18): Just confirming frame numbers
Swordless Link (21:34:35): k
AKA (21:44:27): Now to get myself something to eat :y
Swordless Link (21:44:55): I'm still stuffed from dinner
AKA (22:09:54): Good luck on testing
Swordless Link (22:11:06): Thanks, *actually begins now*
Swordless Link (22:20:24): FUCK YES
Swordless Link (22:20:24): YES
Swordless Link (22:20:30): FUCKING YESSSSSSSSSSSSSS
Swordless Link (22:20:31): I'VE DONE IT
Swordless Link (22:20:32): I'VE DONE IT!
Swordless Link (22:20:35): I'VE GOT THROUGH!
Swordless Link (22:20:56): WE'VE DONE IT AGAIN, MY FRIEND
Swordless Link (22:21:22): I didn't even use the pillar!
Swordless Link (22:28:47): Dammit, where are you? =(
Swordless Link (23:17:47): God dammit
Swordless Link signed off at 23:17:50.
Swordless Link signed on at 23:17:59.
Swordless Link signed off at 23:18:50.
AKA (23:18:58): What have you done?
Swordless Link signed on at 23:19:48
AKA (23:19:57): What have you done?
Swordless Link (23:20:10): Observe, and be amazed, it's HUGE
AKA (23:20:35): Can't wait to see this
FILE TRANSFER
AKA (23:21:56): Fuck me
Swordless Link (23:22:41): You found the method for 1 star, I found the method for 0 star, there's nothing we can't do! ^_^
AKA (23:23:18): We can't theoretically improve the game anymore, apart from opening the door in the moat, which is impossible.

With thanks to...

Myles Bukrim: For supporting the run and checking parts of it regularly to see if there was anything which could be improved.
mr_roberts_z and Rikku: Their previous run served as a good guide for this one, as theirs was about as optimal as a run which follows that route can get.
laughing_gas: For telling us that diving is definitely faster than long jumping, which allowed us to save quite a number of frames.

adelikat: Accepting for publication as an improvement to the 1-star movie.

1 2
6 7
Former player
Joined: 12/27/2006
Posts: 532
Location: Göteborg, Sweden
Bisqwit wrote:
Goes to show that the ways, that the encoder guideline "the tool-assisted disclaimer should overlap prominently enough in both the position and the moment that removing it would be too much work (though not mask out anything particularly interesting)" is followed, can still have some improving in efficiency...
I believe this guy encoded it himself from the .m64, notice the white line at the bottom and the text appearing at the title screen.
My published movies [03:45:05] <Naohiro19> Soulrivers: ... [03:45:19] <Soulrivers> ? [03:46:35] <Naohiro19> <Soulrivers> No! <Naohiro19> So? <Soulrivers> Yes! [03:46:48] <Naohiro19> joke
Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Seconding what Soulrivers just mentioned, I downloaded this elsewhere before the AVI was published on this site, but after the .m64 was submitted. My version does not have the disclaimer. I'm downloading the "real" one now. The guy probably encoded it himself, as mentioned.
Joined: 1/2/2008
Posts: 1
o my gosh.... thats just incredible! wow.... just plain.... wow
Mitjitsu
He/Him
Banned User, Experienced player (532)
Joined: 4/24/2006
Posts: 2997
About the door in the moat outside the castle, I pretty much declare it impossible to open. Charging through the surface at hyperspeed regardless of how fast makes no difference, He always treat it like a swimming envoirment after a couple of frames, even if you could walk on the bottom (using metalcap) it makes no difference since the door does not allow it self to be open.
Joined: 11/15/2004
Posts: 804
Location: Canada
Holy crap, how did I miss this? I mean, I was pretty busy in December, but I still checked in once in a while. I can't believe I'm just learning about a 0 star run! The camera was a bit too close to Mario in Bowser 1 for my liking, and that's really the only thing that bothered me. Everything from the camera to Mario's antics was perfect for the rest of the game. BLJing from Bowser 1 to the basement was awesome. Skipping DDD was brilliant (I knew it could be done!). Actually, I was wondering what was going to happen when Mario jumped out of the level. He just cleared the "DDD barrier" on the way out. It would have been interesting if Mario had landed behind it. The jumping in Bowser 3 is just outstanding. A work of art, really. I can't believe that I missed the TAS event of 2007.
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Mitjitsu
He/Him
Banned User, Experienced player (532)
Joined: 4/24/2006
Posts: 2997
hopper wrote:
The camera was a bit too close to Mario in Bowser 1 for my liking, and that's really the only thing that bothered me. Everything from the camera to Mario's antics was perfect for the rest of the game.
I take it you're talking about when he does the blj. Sadly, I don't think we could have possibly have choosen a better camera angle, If we were to use the Lakitu cam, then it can't keep up with Mario properly. It starts to pan and spin and it would make it difficult for the TASer and the viewer. Plus Mario gets obscured in a few sections.
1 2
6 7