Post subject: How exactly are glitches done in a TAS?
Joined: 5/13/2013
Posts: 180
I know it may sound like a really dumb question (mainly because I'm not as skilled or trained in TAS as many people are on this site) but I've seen a lot of TASes with Heavy Glitch Abuse (the one that REALLY got me intrigued was the one of Zelda 2 where Link was bouncing everywhere like a sweaty dodge ball) and I'm intrigued. I don't really know if it involves editing addresses or anything or just simply doing something in the game, but if anyone's got any tips, please let me know. Thanks in advance!!
A wise man once said "Damn, that's one hell of a steak."
Warepire
He/Him
Editor
Joined: 3/2/2010
Posts: 2178
Location: A little to the left of nowhere (Sweden)
Everything you see are just controller input and sometimes resets of the console, what makes the glitches happen is just timing everything really well, which is accomplished by using re-recording and frame advance, and sometimes also LUA bots. An actual TASer can probably explain in more detail how this works.
Editor, Skilled player (1439)
Joined: 3/31/2010
Posts: 2108
There's a lot of different things that can go wrong in a game, usually, a glitch is caused by some problem of interlocking game mechanics or an oversight by the programmers. Often, it's a combination of mechanics the game designers didn't expect that causes a glitch. You really have to look at each game individually to see what you can exploit in there, and some games offer more room to glitches than others. There's a few common cases that appear in multiple games, such as damage boosts or entering walls to initiate zips in platformers, so it's always worth checking for those. As an example: The flagpole glitch in Super Mario Bros. is caused by entering a solid block by abusing the primitive collision and wall ejection system the game uses: You enter the block at high speed with the correct subpixel position at the top left corner and on the frame the game wants to push you out of the wall again, press Left on the d-pad to get pulled in the other direction instead. If you're far enough in, you'll trigger the flagpole glitch.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
There's no cheating of any kind involved; all the glitches used in TASes can be done on actual consoles if you press the buttons fast enough. It can be instructive to watch console runs of something like Ocarina of Time; the console players get their glitches frame-perfect via spamming the pause button in order to create lag, slowing the game down enough that they can get all the inputs in at exactly the right moments. Common causes of hyper-speed glitches involve doing something on the first possible frame (in many platformers, if you jump the exact instant you hit the ground, you won't lose speed, meaning you can often maintain a faster-than-normal speed), and messing up a game's wall ejection routines (turn round the instant you hit a wall, and many games will believe you to have entered it from the other side and eject you backwards, pulling you deeper into the wall; and wall ejection is often very fast). Many glitches are actually discovered by players just playing on console, and eventually get added to the TAS, in fact. Sometimes they're found specifically for the TAS, but that's rare (except in games which are very heavily studied).
Active player (328)
Joined: 2/23/2005
Posts: 786
All glitches are basically discovered on accident. They could be discovered while playing a game normally, or by intentionally trying to confuse the game and watching what happens. When a glitch is discovered, it can be researched, experimented with, and abused to its maximum potential with a TAS. Pushing a glitch to its maximum potential often involves frame-perfect input and memory watching to figure out just how it works. Many "zipping" glitches (Like the ones in Zelda 2 and Battle of Olympus) are due to pressing Left+Right on a controller on the same frame, something that has unpredictable results because many programmers didn't account for it, because it was assumed impossible to do. Yet, it is possible through either breaking a control pad, or using different types of control pads (L+R can be done surprisingly frequently when playing NES Virtual Console games with a Wii remote), or by just using a keyboard.
Site Admin, Skilled player (1254)
Joined: 4/17/2010
Posts: 11475
Location: Lake Char­gogg­a­gogg­man­chaugg­a­gogg­chau­bun­a­gung­a­maugg
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
CtrlAltDestroy wrote:
All glitches are basically discovered on accident.
I would argue that at least some glitches have been found by examining the game's machine code.
Personman
Other
Joined: 4/20/2008
Posts: 465
Warp wrote:
CtrlAltDestroy wrote:
All glitches are basically discovered on accident.
I would argue that at least some glitches have been found by examining the game's machine code.
And lots more are found on purpose by methodical searching or logical reasoning about how systems might interact in unusual circumstances. Saying all glitches are discovered by accident is not just untrue, it kinda pointlessly undercuts all the hard work that people put into glitch hunting.
A warb degombs the brangy. Your gitch zanks and leils the warb.
Active player (328)
Joined: 2/23/2005
Posts: 786
Here I was using the word "accident" to mean "unexpected". If a glitch is found, then it is certain that something unexpected happened to a player at some time. It is true that you can logically deduce the nature of a glitch, but before you do that, you need to find evidence that a glitch could exist, and that is found by gaining a thorough understanding of how a game is supposed to work, then waiting for (or trying to cause) an unexpected result. For instance, the Pokémon Yellow movies were made with lots of hard work and understanding of assembly programming, but it was all possible because one day a TASer was watching the game's RAM and found something unexpected happen when they reset the game after saving.
Mitjitsu
He/Him
Banned User
Joined: 4/24/2006
Posts: 2997
Shortcuts are often found by skill using various forms of trial an error and discussion. Glitches are nearly always found by fluke or experimentation.
Experienced player (588)
Joined: 2/5/2011
Posts: 1417
Location: France
One of the best ways are to try, try, try, try. For the Zelda glitch, it was L+R
Current: Rayman 3 maybe? idk xD Paused: N64 Rayman 2 (with Funnyhair) GBA SMA 4 : E Reader (With TehSeven) TASVideos is like a quicksand, you get in, but you cannot quit the sand
Patashu
He/Him
Joined: 10/2/2005
Posts: 4043
A lot of glitches happen because you have two things happen on the same frame, and their side effects are processed out of order (many glitches in SMW for example)
My Chiptune music, made in Famitracker: http://soundcloud.com/patashu My twitch. I stream mostly shmups & rhythm games http://twitch.tv/patashu My youtube, again shmups and rhythm games and misc stuff: http://youtube.com/user/patashu