Some time ago, it was established that power-up incrementation could only be done with a Yoshi. So if no Yoshi is present, then it would not be possible to get a Cape Feather out of a Fire Flower. But with the advent of arbitrary code execution, does anyone know if it's now possible to get a Yoshi or Cape Feather in a level where one isn't given?
I'm guessing it's still not possible because such tricks still involve a Yoshi, but I'd love to be proven wrong.
Snes9x 1.53 is still buggy in some aspects. For example, rapid loading and creation of save states often causes the emulator to freeze, and in one case even corrupted my SRAM file.
Part of my current job involves running automated test cases on Android applications. We use a program (similar to Robotium) that gets information about the active UI elements on the screen via instrumentation and sends commands accordingly.
This got me thinking: would it be possible to use these tools to TAS Android games? Has anyone here ever looked into this?
On a similar note, would it be possible to use a program like Selenium to TAS HTML5 games?
Sorry if this is an old topic, but does anyone know if there are any tricks for obtaining a cape feather if none are provided?
Such a trick likely wouldn't be too beneficial for Super Mario World as it's fairly easy to obtain a cape, but it could be useful for TAS'ing hacks that do not provide feathers at all.
If you could play any video game "perfectly" (like in a TAS) in real time, which game would it be, and why?
For me, it would be Super Mario World, which has always been one of my favorite games. The game's unique mechanics allow for many awe-inspiring tricks - not to mention the plenty of opportunities to show off!
A similar concept is already in use elsewhere. Some RuneScape (a popular MMORPG) players automate their in-game tasks with a "smart" macro called SCAR. The program tries to recognize in-game objects by detecting patterns on the screen. It then uses a script to determine where to click next.
The RuneScape policies actually prohibit the use of bots and macros, but such a program could be useful in other browser-based games. The only problem is that the run time probably won't be the same on every computer.
The "tricks" page says that releasing → when Mario is in mid-air will make him travel at his current until he reaches the ground. However, when I release → at the desired speed, it will change two more times before being locked in.
For example, Mario's sprinting speed oscillates in a cycle of 48-47-48-47-49. However, if I release → at 49, Mario speed will change two more times, stopping at 47.
Am I doing something wrong?
I've noticed that some YouTube users are pretty good at making tool-assisted speedruns. Does anyone know if any of the following folks are members of this forum?
penangbenny - I know he's definitely aware of this site because has uploaded one of TASvideos' movies to his account (I believe he also goes by PangaeaPanga on SMW Central)