What a pleasant surprise! A pleasure to see this game demolished with ACE. Yes vote!
I'd love to see any% no major glitches and 100% improved too, but I understand those will be much bigger undertakings.
This discussion reminded me of an idea that I think has been thrown in the past. Why hasn't anybody created a game that's quite directly based on TASing?
I remember seeing such a game posted on the forums in the past - I can't remember what the game was called, but the premise was that you had a 'car' with rockets on both ends, gave you a scrubbable timeline you could add/change inputs on arbitrarily, and it showed the predicted trajectory of your current inputs. It was called something like rctr ('rocket car').
Trick 4: Air Jumping. When Hardy walks off a platform instead of jumping off...he can jump during the fall while still in mid-air. This is utilized a few times for better jumping positions and to make jumps that wouldn't otherwise be possible. Again, likely a programming oversight.
This is a common mechanic in platformers, introduced whenever the game designer wants jumping to have more lenient windows, and it's called 'coyote time'. (Although using it to make otherwise impossible jumps could be an oversight, especially if it sequence breaks.)
Another mechanic like this is called 'jump buffering', which allows for jump to be inputted before you land, and to execute on the first possible frame.
810 note door skip and oot any% improvement on the same day! What a remarkable achievement, nice work! If it becomes RTA viable even better, good luck!
'I spent hours trying to install Shockwave plugins for modern browsers. None of them run the game well at all. The game freezes or crashes randomly.'
Dumb question, but why not install Shockwave for a NON modern browser?
It's a delight to see one of my formative TASing experiences revisited and made more powerful than ever. This game being so deliciously and intricately broken on both the high level of creating objects from nothing, permanently glitching the graphics and the low level of just zipping, wrapping and pause buffering through every obstacle in your path made it a defining TAS of its era. Yes vote, and I look forward to the no object glitch TAS as well!
Thank you for TASing the game end glitch! This turned out great: complex collection routing, platformer mechanics, cool speed glitches and a surprise finish. Yes vote.
Thank you! This TAS was a joy to experience, between the extremely detailed submission text that goes into not just the final product but the evolution of the TAS, a game I have a ton of fond memories of and the TAS itself. All the effort you've put into every aspect of its presentation has paid off.
Favourite moments:
* Buzz fight, the beautiful synchronized dance
* The rapid reverse traversal of Bamboo Terrace and Spooky Swamp
* The TAS precision massacre of Sgt. Byrd's Base
* The LAUNCH in Ice Peak, and the detailed description of the weird glitchiness behind it
* The rapid dismantling of Spike
* Lost Fleet, both the proxies and the abuse and quick re-entry of zombie mode
* Bentley's Outpost, going fast as a yeti by exploiting zombie mode teleports
* The huge launch in Charmed Ridge
* The utter absurdity that was the Sorceress battle, and its detailed description
Yes vote, very amusing. It's interesting to think about how many games might have an out of bounds trigger just waiting to be discovered like this one.
It actually really isn't that hard to beat in TAS, considering the player character can be made invincible at any time (which the TAS very liberally makes use of)
This is correct, and it is mostly not hard as a result. However, there's 1 frame between piledrives (4 frames between counters) where you are vulnerable, and this makes dodging some of Mecha Santa's attacks non-trivial, since they fill the screen THAT much. Notably, a previous person tried to make a TAS of it for the hell of it, and got stuck (as in, couldn't finish) on Mecha Santa.
There is a lot of shared code, particularly all of the Koei games have the built-in virtual machine + a standard C library in their Root page, so it is a lot faster going now that I worked most of it out already.
how did you figure that out? Is this information available somewhere? I'd be interested to see a write up on it.
