(Link to video)
I found a trick right after my “all instrument” run was finished that boosted link several pixels, but I wasn’t sure how I specifically did it and if it worked on the DX version. This run started when I verified that the trick worked and worked its way from there.
  • Emulator used: VisualBoyAdvance re-recording v22
  • Aims for fastest time
  • Abuses glitches
  • Manipulates luck
  • Takes damage to save time

Tricks used

Up/down + A or B combos

This allows link to do a variety of glitchy things with items such as the sword, shield, shovel, etc… Using up/down/b allows the owl text near the sword to be skipped.

Corner Boosts

When you walk past a corner you can move into it slightly to push Link a maximum of 2 pixels forward. Often you can only save 1 pixel due to the sideways walking mechanics.

Screen Transition Boosts

This is the new trick I found. First off, this trick only works when Link is traveling vertically between screens. By clipping as close as you can into a wall that leads to the next screen and pressing the direction you’re headed plus the direction of the wall (for 1 or 4 frames depending on diagonal movement mechanics), Link gets a “boost” when he can first start moving on the next screen. This saved around 2-3 pixels. Ideally, you want to just press the direction for 1 frame, because you usually loose a frame if more are held, but it can’t be avoided in some cases. An area where this doesn’t work in this run is after killing the 2 enemies and walking back to the house when exiting the screen with the 2 kids playing since Link can’t clip the few pixels into the wall getting him as close as possible.

Useful memory addresses

0000FF9A – x speed (in subpixels)
0000FF9B – y speed (in subpixels)
0000FF98 – x coordinate (in pixels)
0000FF99 – y coordinate (in pixels)
0000DBB5 – number of enemies killed in the area below town.

Improvement list

Obtaining the Sword

-Using a transition boost while entering the screen with the kids playing saved 1 frame. (1 saved)
-Using another transition boost saved 3 more frames. (4 saved)
-Lost 1 frame right before entering the screen with the first damage boost (either from screen loading lag or subpixels). (3 saved)
-Saved 3 frames by moving on certain frames while pushing the enemy. (6 saved)
-Lost 1 frame during the sword text? (5 saved)
-Lost 1 frame while bringing up the save menu from a frame rule (I tested delaying finishing the text by both 1 and 2 frames and it ended up being the same time, but waiting 3 frames lost 4 frames). (4 saved)

Stealing stuff and setting up the doghouse glitch

-Saved 1 frame stealing the bow from a better shopkeeper cycle or something. (5 saved)
-Saved 10 frames boosting off Bow-wow. Link was 1 extra pixel higher since this was the best manipulation I could get during the Bow-wow boost. I don't actually know if this lost time due to how I'm comparing the runs. Also manipulating bow-wow was a pain. I resorted to testing random shield inputs on the prior screen until I got movement that looked decent, and worked off that to find if it was. (15 saved)
-Saved 1 frame due to a transition boost. (16 saved)
-Saved 2 more frames from another transition boost. (18 saved)
-Saved 3 frames killing the enemies since the lower enemy starting moving towards the left sooner (the only reason I can think of why this happened is since I got there faster). (21 saved)
-Saved 5 frames due to a transition boost and possibly less loading lag. (26 saved)
-Saved 5 frames by shield bouncing off Bow-wow (with <v>) after the slash boosts. (31 saved)

Getting to the Boss

-Saved 4 frames by having better randomness on the glitchy tiles on the screen where the little Bow-wow lives. I believe the randomness had to do with Link's subpixel values which I believe changed a bit during the shield boost. (35 saved)
-Saved 1 frame on the next screen, not sure how. (36 saved)
-Saved 4 frames on the screen where the tiles start by pushing link backwards a bit. No idea why I saved some here either. (40 saved)
-Saved 1 frame right before entering the boss battle from having better pixel placement. (41 saved)

The good ol’ Boss fight

-Saved 3 frames on the "blob" form of the boss for some reason. (44 saved)
-Saved 1 frame after killing the worm form due to better manipulation of where the worm died. (45 saved)
-Saved 1 frame around the 3rd sword attack or so on Ganon somehow. (46 saved)
-Saved 2 frames because the eye form opened 2 frames earlier. (48 saved)
-Saved 8 frames from lucky eye manipulation. This was the part of the run which I wasn’t looking forward to, and with good reason. Manipulating the eye form is so random from what I’ve tested. Anyway, on my first go, somehow I was able to make the eye “blink” after the 1st damage wave. By blink, when the eye closed, it only closed for 1 frame (verses 9-14 or so in the last run). After that I had terrible manipulation and when I finished I was only 44 frames ahead total.
I redid playing with the boss before the eye first opened and got it to close for 13 frames. After that I got lucky and got it to blink which saved some time. It MIGHT? Be possible to get the boss to blink twice, saving like 12 frames, but I had no luck.
Total frames saved was 56, almost a second.
Possible improvements include finding some way to manipulate the eye boss so that it only closes for 1 frame each time. I was unable to do this, but it theoretically should be possible. Another thing that might help improvements is if there is a way to watch Link’s current subpixel address. I tried finding it, but had no luck.

Nach: I was a bit shocked to see such a run, as I don't think there's a place for it on the site. I was even more shocked to see that there are past glitched runs for LADX that have already been accepted. I don't think this branch has a right to exist under our current structure and rules, and don't even know how there was such an oversight when the previous runs were accepted.
The precedent seems to be that if a run is a good improvement to an existing run, and the audience likes it, it is accepted despite the issues. Since this is indeed a nice improvement to an existing run, and the overwhelming majority of the audience seems to want the run, I will be accepting it. However, at some point in the future, we'll be needing to clarify our rules and consolidate these runs into a single branch, or reorganize them according to a new site structure.

Brandon: Publication underway.

Emulator Coder
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
creaothceann wrote:
Just declare them invalid in the rules and don't publish them then?
I actually brought that up back in 2004 or so. I'm sure someone can find the thread. In any case, if we make them invalid, then we can't have our crazy DKC videos, can we?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
creaothceann
He/Him
Editor
Joined: 4/7/2005
Posts: 1874
Location: Germany
turska wrote:
All published revisions are official.
Yes, but some versions are more official than the others. ;) When a game designer looks back at his game, s/he will (imo) consider the latest version the definite one.
turska wrote:
More recent also isn't necessarily less glitched - sometimes game-breaking glitches only exist in the latest revision.
But not the majority?
turska wrote:
There's also funny cases like Commander Keen 1 where 1.34 was only released in the UK, 1.32 was part of a special Gravis deal and the most commonly encountered version is 1.31. What would be the preferable version in such a case?
1.34, I'd say without knowing what was fixed. (EDIT: Seems like no game-changing issues were fixed.)
Nach wrote:
if we make them invalid, then we can't have our crazy DKC videos, can we?
We can, just not as featured videos.
Experienced player (508)
Joined: 11/2/2010
Posts: 359
I believe all versions are valid and that the only qualifier on what version you use should be what makes the run most meet the goal. In this case it is a aims for fastest time while using glitches run, so you should use whichever version of the game that allows you to meet that goal. The GB achieves that goal better than the GBC version so that is the version that should be published. If the goal was speed but GBC could finish earlier than that version should be used. Conversely if the run was just for entertainment, than you should decide which version would allow to maximize entertainment. In this case it sounds like everyone believes the GBC version is more entertaining so this argument should have occurred back before the playaround was published since it doesn't meet the goal to it's fullest.
Joined: 7/2/2007
Posts: 3960
goofydylan: the question really is, are the GB and GBC versions separate games? The only gameplay difference I'm aware of is that the GBC version adds a bonus dungeon, though people who have actually played the games may be familiar with more differences. If we were just dealing with e.g. the 1.0, 1.1, 1.2 versions of a given game, then I agree with your reasoning. But the question is if GB Link's Awakening is a different game than GBC Link's Awakening.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced player (508)
Joined: 11/2/2010
Posts: 359
Derakon wrote:
goofydylan: the question really is, are the GB and GBC versions separate games? The only gameplay difference I'm aware of is that the GBC version adds a bonus dungeon, though people who have actually played the games may be familiar with more differences. If we were just dealing with e.g. the 1.0, 1.1, 1.2 versions of a given game, then I agree with your reasoning. But the question is if GB Link's Awakening is a different game than GBC Link's Awakening.
I feel like in this case they are the same game and am making my argument that way. If the only gameplay difference is a bonus dungeon, which isn't seen in this run, then why say they are different games. Look at both of them at the same time. Here is how it goes. 0-38 seconds: Identical except DX is in color and text is different making this version slightly slower (aka less entertaining) 38-59 seconds in GB, 38-74 seconds in GBC: Moving to get sword. GB more interesting through use of glitch to get there in less time and abusing the game. 59-85 seconds in GB, 74-153 seconds: Glitching the game to get to the boss fight. Admittedly, these segments are probably equally entertaining in my mind as they show off slightly different segments but I would definitely not say that the GBC version is more entertaining or better. 85-158 seconds in GB, 153-266 seconds in GBC: Both glitched out boss fights. I would give the advantage to GB for sure as it glitches out more thoroughly and there is less text. 158 on in GB, 266 on in GBC: Identical. After breaking it down to the segments I can't see how it can be argued that they are different enough to warrant two publications. Two segments of the run show major differences and the difference is that there is more text and less glitches, but still major glitches, in the GBC version. Other than that you have to watch side by side to see any difference and the GB version actually even wins those segments as there is less text in them.
Emulator Coder
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
goofydylan8 wrote:
Conversely if the run was just for entertainment, than you should decide which version would allow to maximize entertainment. In this case it sounds like everyone believes the GBC version is more entertaining so this argument should have occurred back before the playaround was published since it doesn't meet the goal to it's fullest.
In terms of pure entertainment: For 100%, LADX is better. For playaround, LA is better. That's what we have.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
in case this isn't accepted,at least the players should update the this game's page of tricks and glitches here at SDA.i do think it should be accepted since it's technically another game!
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Player (65)
Joined: 4/21/2011
Posts: 232
Nach wrote:
In terms of pure entertainment: For 100%, LADX is better. For playaround, LA is better. That's what we have.
And that is based on judging, right? IMO too much thought is being put into making an absolute rule for something which should be treated on a case by case basis.
Active player (422)
Joined: 9/27/2004
Posts: 650
Location: Canada
nanogyth wrote:
IMO too much thought is being put into making an absolute rule for something which should be treated on a case by case basis.
I agree! Further, I think in this specific case, both runs are good and sufficiently different in terms of tactics and execution and should both be published. I hesitate to cite precedent, because it's pretty well pointless, you know since past mistakes don't justify whatever that credo is. But I think this is sort of in line with the two Marble Madness runs in terms of watchability. Though maybe a bit closer to the FDS version of Mario Bros being published alongside the NES version in terms of what version of the game is used. I'm not sure if I added anything that I didn't add before, but I used a few more words this time.
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
just think of pokemon green,yellow,red runs and you'll probably reach the conclusion this should get published for its unique methods of glitching to the end
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Joined: 2/25/2006
Posts: 407
creaothceann wrote:
re: revisions I'd say use the latest revision for publications because it's the "official" (most complete) game & shows more of it, and use previous revisions with their glitches for playarounds.
This so damn much, it's the primary reason why I hardly visit here anymore (Aside from the ugly website redesign, though that's just my personal opinion). You're meant to be demonstrating how fast the game can be beaten and doing that with the most glitched up version of the product is just pointless because you just end up with stuff that is less playing and more completely non nonsensical. The TAS's just end up a pile of brokeness with a game name slapped on top. Sure there may be some games where the latest versions can be just as broken but there would be far fewer such cases then there are now where most games is just a mess like look at Super Mario World 2, the game just (seemingly) randomly cuts to the end, how is that entertaining? (Bad example, I know but I'm terrible at coming up with examples.)
Ryzen 3700X, ASUS Crosshair VIII Hero (WiFi) Motherboard, 32GB 3600MHz RAM, MSI Geforce 1070Ti 8GB, Windows 10 Pro x64 http://tasvideos.org/Nach/FranpaAlert.html
AUS
Joined: 4/6/2012
Posts: 27
Nach wrote:
Under the above logic, we should have an unlimited amount of submissions, each using a different route through the game, since you know, they're different gameplay.
I would only agree if they were literally the same game. They are separate games.
Emulator Coder
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
AUS wrote:
Nach wrote:
Under the above logic, we should have an unlimited amount of submissions, each using a different route through the game, since you know, they're different gameplay.
I would only agree if they were literally the same game. They are separate games.
That's a matter of a opinion. The same like I wonder why we have runs for Pokemon Blue and Red, although at least there, we don't use the same goals.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Editor
Joined: 3/31/2010
Posts: 1466
Location: Not playing Puyo Tetris
Link's Awakening and Link's Awakening DX is like Pokemon Red and Pokemon Fire Red. Same game (mostly) but with less bugs in the newer one.
When TAS does Quake 1, SDA will declare war. The Prince doth arrive he doth please.
Joined: 8/23/2008
Posts: 417
What a strange debate that has arisen. I do not envy the judges of this site. Essentially, the question Nach asks has no good answer. Ultimately this is a grey area, and grey areas are anathema to rule-makers. There is this idea that runs that are accepted give some precedent that could be trouble later, but this is only true if the judges do not hold runs to adequate standards. I personally think this run should be published; I also think that SMAS runs ought to be published as well, even if they are simply remakes. There is a certain amount of entertainment value available in publishing a run of the modern, graphically-updated version of a game, and there is also value in having a run of the original version of a game on the site. Remakes are not painfully common; as long as a run well is done well and is interesting to watch, it seems reasonable to allow publications of both originals and remakes here. Ultimately, I have to ask "why not?" There has long lingered this sense of apprehension about overly eager publication, as though allowing too many quality TASes to be published would be harmful to the site. I understand this argument, but I think the risks are over-exaggerated and the gains underrated. As long as the game choice is good and the category interesting, an optimized TAS should have a home here, I think.
I will not use self-reference in my signature.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1022
I'd say with different glitches, it's pretty much different categories. (Using the most recent version of a game, if it contains no known glitches, might lead to an interesting way to do "glitchless" TASes; those are less popular than ones that use glitches, but there's still a large contingent of people who like watching them.) If the game falls to the same glitches used in the same way, it probably isn't interesting.
Experienced player (508)
Joined: 11/2/2010
Posts: 359
ais523 wrote:
I'd say with different glitches, it's pretty much different categories. (Using the most recent version of a game, if it contains no known glitches, might lead to an interesting way to do "glitchless" TASes; those are less popular than ones that use glitches, but there's still a large contingent of people who like watching them.) If the game falls to the same glitches used in the same way, it probably isn't interesting.
If using different glitches constituted a new category then #3592: SwordlessLink's N64 The Legend of Zelda: Ocarina of Time in 22:50.27 would not obsolete [1616] N64 The Legend of Zelda: Ocarina of Time by Bloobiebla in 56:54.20 because they are both pretty glitched but where the new submission uses Ganondoor and the old one doesn't.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1022
Using different glitches can constitute a different category, even within the same game (see, e.g., Super Mario 64 70-star); however, a newer version of the game gives a natural point to draw the line. Generally speaking, "past most-glitched" (e.g. Super Mario 64 16-star) isn't an interesting place to draw the line, or at least hasn't been found to be one here at TASvideos. For low-glitch categories, you're either going to want a lot of different glitches (as in SM64) to make the runs sufficiently different from each other, or else you're going to want no glitches at all to show a true glitchless playthrough.
Active player (490)
Joined: 1/12/2007
Posts: 682
Here's a wacky idea: publish good runs that people enjoy. This run should be published. SM64 70 star should be published. OoT's old any% shouldn't cause it's a boring, confusing fetchquest. IDK what's hard about this or why the "rule" has to apply to every game.
Joined: 8/23/2008
Posts: 417
goofydylan8 wrote:
If using different glitches constituted a new category then #3592: SwordlessLink's N64 The Legend of Zelda: Ocarina of Time in 22:50.27 would not obsolete [1616] N64 The Legend of Zelda: Ocarina of Time by Bloobiebla in 56:54.20 because they are both pretty glitched but where the new submission uses Ganondoor and the old one doesn't.
My only objection to this point is that the glitches done in OoT are both possible in the same game, whereas the glitch done in LA cannot be done in LADX and vice-versa. In other words, it's not quite the same thing to compare an improvement to a trick only doable on a different version. (If there are gameplay differences besides text in OoT that come to light in the two runs you mention, please correct me, as the submission text for SL's JPN-version run itself does not seem to suggest that any relevant version differences exist otherwise).
I will not use self-reference in my signature.
Post subject: Movie published
TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14873
Location: 127.0.0.1
This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [2077] GBC The Legend of Zelda: Link's Awakening DX "warp glitch" by Bobmario511 in 05:37.87
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11267
Location: RU
Oh wow, these 2 Zelda HDs have the shittiest subtitles I've ever seen.
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.