Joined: 8/14/2009
Posts: 4083
Location: The Netherlands
In light of the current discussion on the category name for the recently published VVVVVV run, I think it's time to make a thread to discuss the general issue.
In the past, runs that were considered faster than 'conventional runs' (typically by using heavy memory corruption, or the like) were given the branch "glitched", while the conventional runs had no branch (referred to as "any%").
Several months back, this system was changed - as the definition of "any%" constitutes the fastest possible way to beat a game, the branch "glitched" has effectively been removed, as what used to be called "glitched" is now simply "any%". However, this has led to an issue with categorizing the previously any% runs - as they are not the fastest runs, they needed a branch to be identified as. This has led to run labels for previously any% movies, like "Super Mario World "11 exits"", "Super Metroid "no X-Ray glitch"", "Crash Bandicoot 2 "no box glitch"", the recently relabeled "VVVVVV "no Esc"", and a run that was at one point referred to as "Super Mario World 2: Yoshi's Island "no L+R, no null egg glitch"".
A lot of criticism has been raised over this categorization (primarily on IRC earlier on, but now moreso in the forums as well), as the branch names can be confusing or nondescriptive for newcomers. I keep seeing more and more people who want the label "glitched" brought back and the definition of "any%" on the site reverted to before the change.
On the one hand, "glitched" is clearer for audience viewers; on the other hand, it is arbitrary, and leaves an arbitrary definition for "any%". On the other hand, "no x glitch" is an objective definition, and provides an objective definition for "any%" (fastest run possible) but can lead to long or confusing branch names.
So I'll put this poll here - which label system is preferred? Should we bring the "glitched" label back, or retain the current system as it is now?
Related discussion topics:
VVVVVV branch discussion (the currently ongoing issue)
"Movies labeled "glitched" that shouldn't be"
http://www.youtube.com/Noxxa
<dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects.
<Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits
<adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
My quote from VVVVVV:
In a long run, we would need a system that makes, say, 5 branches per game easy to handle on both the user and the staff sides. With Moons, you never know which branch will appear next, and if there are plenty, the system must be crystal clear. For now, any solution for that was delayed until there are actual problems. And only a few games provide real difficulties in categorization. But as I said, Moons remove the branch limit, so if we don't solve it in time globally, it then gives us more problems to solve locally, just like this one.
First, I want to state, that "glitched" itself is as ambiguous as "any%": because it's literally anything that could lead to fastest/glitched game completion. Super Metroid is a good indication of that: both 2 fastest runs are "glitched" by what that traditionally used to mean, but they are both so well-received that they're both published. You can't get away with naming them "glitched". You would have to state, which glitch they use, or, maybe, which kind of a glitch it is. So in a long run, "glitched" label, though it showcases the difference from the "less glitched" branches, isn't unique enough to allow SEVERAL glitched branches of the same game, if they are well-received.
Then, I would like to show my personal preference, that accounts for the long run of the Moons system and removes ambiguity from "any%". Here it is:
No label must be left only for games that have no variety on how to complete it. If it gives you several endings, each ending must be labeled. If you can choose the amount of players, or the character that alters the gameplay over the other(s), put it as a label. If it is warps, their use must be part of the label.
That way, the viewer will in 100% cases know what he's looking at. Basically, it's about which movie tags must become branches. Some of them already are, some better not be.
If such approach is applied to varieties of glitched branches, there may appear "game end glitch", "SRAM glitch", etc. that would work for quite some cases, showing the similarity.
This is all because I don't see any benefits in the "drop the label for every fastest branch" approach. It simplifies some management, but makes the other things much harder. And when used along with "no X" thing, it spawns really ugly issues.
Also, I realized how to resolve the difference between the glitched branch and heavy glitch abuse tag. People always confuse these 2 concepts, since they are both sequence breaking, so here's the solution:
Heavy glitch abuse is when one does sequence breaking by some in-game means (game engine faults). Soft sequence breaking. Glitched is when one corrupts the gameplay and breaks the logical conditions for common soft sequence breaking, overcoming the gameplay limitations. Hard sequence breaking. What do you think?
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.
It's worth noting that VVVVVV's so-called "glitched" rundoesn't use memory corruption, arbitrary code execution, or anything of the sort; it instead exploits a straightforward bug in the code.
The advantage to the current system is that all branch labels are objective and clear. A run called "no warps" doesn't use warps; a run named "all gems" collects all gems in the game; and a run labeled "no breadcrumbs glitch" doesn't use the breadcrumbs glitch. This has worked fine for 99% of the 2500+ runs on the site.
The problem with "glitched" vs "non-glitched", and the reason why this was removed in the past, is that nobody so far has given an objective and non-arbitrary definition for it. For some people the distinction is about skipping the final boss, for some it means warping past most of the game, sometimes it's about wall zipping, etc, and this has led to a lot of debate. If we could all agree on a meaning (e.g. defining "glitched" to mean "memory corruption") then that would prevent a lot of future debate.
It strikes me that the underlying issue is not so much the terminology, but the fact that for a handful of games (notably Super Mario World and Super Metroid) there are many different branches, and it is not obvious how these varied goals interact with arbitrary code execution. I think it would be better to consider how to deal with many-branched games, than to vote on which term to use for one of them.
I find option 2 ("glitched" & any%) far too inconsistent since it makes runs like LttP be titled as "glitched" but not runs that look far more glitchier liketheseforexample be titled as such. And before anyone argues that they don't skip bosses or whatever, all those movies linked basically destroy the game such as skipping most of the game.
Edit:
For some people the distinction is about skipping the final boss, for some it means warping past most of the game, sometimes it's about wall zipping, etc, and this has led to a lot of debate.
This movie has all 3 yet I'm pretty certain it won't be considered "glitched".
Let's try out my above approach to solve things :)
Radiant wrote:
It's worth noting that VVVVVV's so-called "glitched" rundoesn't use memory corruption, arbitrary code execution, or anything of the sort; it instead exploits a straightforward bug in the code.
It still uses some glitch, and does a sequence break straight to ending. Right? I'd call it "game end glitch". And the one that doesn't do it may be unlabeled (since it doesn't do anything special while completing the game).
Radiant wrote:
The advantage to the current system is that all branch labels are objective and clear. A run called "no warps" doesn't use warps; a run named "all gems" collects all gems in the game; and a run labeled "no breadcrumbs glitch" doesn't use the breadcrumbs glitch. This has worked fine for 99% of the 2500+ runs on the site.
The problem with that was (and still is) that some branches that avoid the "breadcrumbs glitch" are labeled as "no breadcrumbs glitch", while some others that avoid it (or where it's not possible, but still) somehow don't have that label. Inconsistency. And in long-terms, it's also very confusing for a first-time viewer, because one would assume branches that don't have the "no breadcrumbs glitch" label do use that glitch.
Radiant wrote:
The problem with "glitched" vs "non-glitched", and the reason why this was removed in the past, is that nobody so far has given an objective and non-arbitrary definition for it. For some people the distinction is about skipping the final boss, for some it means warping past most of the game, sometimes it's about wall zipping, etc, and this has led to a lot of debate. If we could all agree on a meaning (e.g. defining "glitched" to mean "memory corruption") then that would prevent a lot of future debate.
Check out my above definition. Even though I'm not for the "glitched" label, but for abstracting the kind of hard sequence breaks into several common labels instead on just one.
Radiant wrote:
It strikes me that the underlying issue is not so much the terminology, but the fact that for a handful of games (notably Super Mario World and Super Metroid) there are many different branches, and it is not obvious how these varied goals interact with arbitrary code execution. I think it would be better to consider how to deal with many-branched games, than to vote on which term to use for one of them.
With Moons, we must be virtually ready for all games having 5+branches. No one now restricts the amount of branches, so what happens now with these few games will definitely happen to others.
jlun2 wrote:
I find option 2 ("glitched" & any%) far too inconsistent since it makes runs like LttP be titled as "glitched" but not runs that look far more glitchier liketheseforexample be titled as such. And before anyone argues that they don't skip bosses or whatever, all those movies linked basically destroy the game such as skipping most of the game.
Yeah, refer to my above solution and tell what you think of it. Though, as I said, I'm not for putting all hard sequence breaks under one label, but rather for grouping them by the type of that.
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.
It's worth noting that VVVVVV's so-called "glitched" rundoesn't use memory corruption, arbitrary code execution, or anything of the sort; it instead exploits a straightforward bug in the code.
It still uses some glitch, and does a sequence break straight to ending. Right? I'd call it "game end glitch". And the one that doesn't do it may be unlabeled (since it doesn't do anything special while completing the game).
Yes it does. It uses dying during the intro to break out of it, and later uses the intro cinematic to overwrite the go-to-gravitron game state. It still sequence breaks the game, using the exact same function (getting the gamestate to a value it shouldn't be in) except it does this in a slightly different way.
It's worth noting that VVVVVV's so-called "glitched" rundoesn't use memory corruption, arbitrary code execution, or anything of the sort; it instead exploits a straightforward bug in the code.
It still uses some glitch, and does a sequence break straight to ending. Right? I'd call it "game end glitch". And the one that doesn't do it may be unlabeled (since it doesn't do anything special while completing the game).
Yes it does. It uses dying during the intro to break out of it, and later uses the intro cinematic to overwrite the go-to-gravitron game state. It still sequence breaks the game, using the exact same function (getting the gamestate to a value it shouldn't be in) except it does this in a slightly different way.
Please elaborate on the difference between 2 runs then. I now am not sure what's so conceptually different about them other than "showing more/less of the game".
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.
Joined: 10/12/2011
Posts: 6337
Location: The land down under.
YAY loosing sleep! YAY Quote chain. YAY no seriously I need sleep.
------------------Part 1------------------
Spikestuff wrote:
Masterjun wrote:
What makes it not properly defined?
Basically it's saying a function which it's not doing and for another run it doesn't mean that it could be used next time to obsolete.
It's easier using something that would make more sense by having a better meaning to the branch then going "No Esc".
It's much harder to clarify a branch name nowadays because we removed "glitched" branch which made it much easier to place certain runs.
An offender to this is Crash 2, we now have the any% run titled "no box glitch" you would only understand this if you play or have an understanding of Crash 2.
To note: the original branch name to that was "no memory corruption"
So in other words we need to cater the viewer (who has minimum knowledge on that game) a branch name that is sensible and understandable.
Basically it's saying a function which it's not doing and for another run it doesn't mean that it could be used next time to obsolete.
Error: not enough grammar.
Yea that's the issue of trying to prove a point so late at night that the points don't appear as you want it.
I'll try and paraphrase
- Branch with "No Esc"
- Next run to obsolete
- Uses Esc to save all crew and skip stuff
- Uh.....
Another way of looking at it is like how we have the 1 and 2player TASes.
When one is faster than the other the slower one will mention how many players.
*yawns* Lets see... Uh the point, the point, right...
So calling something "No Esc" and then having a run obsolete it with Escape would put us back where we started trying to figure out a proper branch name.
Note: This is all in theory.
Scepheo wrote:
"No ESC" is simple, concise and correct.
I beg to differ...
feos' point on the branch name was quite clear. Heck I even agree with his 1 word response to the naming of the non-glitched TAS (Note: I'm writing non-glitched here because it's easier to mention what it was originally known to be on TASVideos).
feos wrote:
wut
If someone want to figure out what I wrote go nuts. Note: It's against "No ESC" as a branch.
(man I love my notes)
Anyways branching it as "No ESC" should be a 2nd last resort and "No Storage" is our last resort.
In Cave Story it's not the worst ending (escape on the dragon) or the intermediate ending (kill the Undead Core) but the best ending (kill Balos).
We don't currently have an any% run of Cave Story, I'm not sure what you're getting at.
Patashu I have to agree, you didn't explain Cave Story properly...
------
Hmm... I think I can use this to drop the branch bomb by keeping only 3 (glitched/any%/100%).
Okay. I'll create a quick glitched/any%/100% list myself off 2 characters.
Crash Bandicoot and Spyro the Dragon.
Even though we have not done any complete 3 TASes on it yet I will still bring this up.(This is mainly due to emulation fault or other)
All these games have their own routes no mater what you try.
Crash Bandicoot 2.glitched, *any% and 100%.
Simple
Crash Bandicoot Warped.
**glitched, *any% and 105%
Spyro 2: Ripto's Rage(Gateway to Glimmer for Europeans (if confused))
*** glitched(speedrunners dubbed this any%), *any%(speedrunners dubbed this 14 tailsman) and 100%
Where can we go with this?
Well there is one game speedrunners themselves don't want to do.
Spyro: Enter the Dragonfly
*any%, *any% ***NMS and 100%
Now before you jump to conclusions to what I just wrote and claim the any% to be a non-glitched branch... it is.
1:07.23 - NTSC Gamecube
* Would not be added into the branch field.
** This is a speedrun just to prove it that there is such thing.
*** No Major Skips
The point of any% is to show the complete difference there is to "glitched" and "100%" they all have their own routes, their own struggles.
glitched: Doing the same damn game skipping glitch over and over again. Stares at Super Mario World - 5 DIFFERENT BRACHES FOR THIS GAME! (could be worse we could have all the SM64)[/size]
any%: Completing as fast as possible with minimum requirements. stares at Metroid
100%: Max requirements, more planning, more routing, more backup strats. unable to stare at anything... there isn't much.(Note on this point I just made: I'm not taking to shit on Metroid or SMW I'm just pointing them out as those 2 games are more acknowledged to me seeing several different versions)
^If you're going to point something wrong on the third one go back to the first one.
Yes I'm seriously going to put you through a loop.... go back to the first part if you disagree on the third then go to second then go to third and keep doing it.
Over and over again, until you understand... or when someone translates the way I write for you. (Possibly Guga or feos)
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account.Something better for yourself and also others.
Please elaborate on the difference between 2 runs then. I now am not sure what's so conceptually different about them other than "showing more/less of the game".
The old (long) run only overwrited the game state by accessing the intro cinematic. This allows you to skip only the gravitron.
The new run uses text storage to alter the game state. This gives more control (although still limited, you can only increment it certain amounts), and allows you to skip to the ending cinematic. This is only possible due the the sheer luck that teleporting (= 4058) and ending cinematic (= 4080) states are close to eachother.
The main difference as such is the use of text storage through going back to the menu. Because most real-time runners dislike doing this, both are considered different branches.
As such, we got to describing the branch. "No Menu" was suggested, but during testing for the run, I discovered (by being told) that it is possible to get text storage without exiting to the menu. This, however, makes it a fairly useless glitch, as it leaves you (as far as we know) incapable of actually accessing any transitions during which you could alter the game state. I also discovered, however, that it is possible to use text storage without actually ever exiting to the menu, except this means you have to manipulate the game state starting at 0. It would take a lot longer, but the combination of these two allow for the end-sequence warp without ever going back to the menu, I believe. As such, "No Esc" was suggested, as that does disable this option (although you select "No" when given the prompt, you do still press ESC to get the "Do you want to go back to the menu" screen).
As such, although the difference in what you see of the game is enormous, both runs glitch the game in the same spirit.
My own opinion on the matter for which this thread was created is that we should not go back to "glitched" as a branch name. This for various reasons:
1) It suggests there is something wrong (in the cheating sense) with the movie, as we clearly find that it should not serve as the main (= unlabeled) movie for the game.
2) When watching the other, non-glitched run, there is often a note in the publication that states there is a faster movie. Apart from this, there is no indication that the run actually is known to be suboptimal. I believe that if a run is purposely done slower than possible, that should be noted.
3) As stated before, nobody (even among those who want "glitched" to come back) agree on when it should be used. I'm not going to elaborate on this further, jlun2 made this point already.
4) These "glitched" runs have in some cases been console verified. They don't use any external modding or anything like that, and have been proven to work on the actual console, for real. Yet, labeling (and thus seperating) them makes it seem as if we suddenly started taking all the "You're cheating" and "That doesn't actually complete the game!" YouTube comments seriously. To me, it feels like we're admitting that we're cheating.
5) An argument often used against "No X glitch" is that it is not a clear description unless you are familiar with the game. However, "glitched" is never a clear description, even if you are familiar with the game.
Conclusion 1
From this last point I draw the conclusion that everybody who supports bringing back "glitched" is okay with unclear labels.
Conclusion 2
The argument against "No X glitch" is that is leads to cumbersome and overly long branch names. From this I conclude that these people want short and one-size-fits-all label names.
Conclusion 3
People who are against "glitched" indicate that it labels the wrong movie. From this I concluse that the fastest run should not be labeled.
Taking these three conclusions, I suggest we introduce the label "Forgoes time-saving glitches" (name up for discussion) and simply slap this on all the runs it applies to.
TASes should be branched individually using the game's (speedrunning) terms and tagged with general things.
When a new TAS makes uses of a new trick, the previous (even if non-obsoleted) movie's tag should contain an easily understandable expression referring to not using a time saver trick(s) that wasn't discovered back then.
I think multiple "glitched" movies should be distinguished by adding the unused techniques to the previous (even if non-obsoleted) movie's branch.
I don't think we need to categorize "glitch" levels. It would get worse when detailing Super Metroid corruptions. The branch should contain the most important difference between the two movies. And I would write an example if I would know which movies counts in this debate.
edit: Regarding branch name length. The various ways to complete a game is what I call the "branch" of the TAS. Therefore if a game allows to be completed using a complex route, the branch's name should reflect it as accurately as possible and should be renamed afterwards if the term gets "obsoleted".
PhD in TASing 🎓 speedrun enthusiast ❤🚷🔥 white hat hacker ▓ black box tester ░ censorships and rules...
Specific vs. generic
For the first issue, I have previously argued that while I sympathize with the wish to have as objective and explicit category names as possible, being too specific runs the risk of overfitting, where one ends up describing incidental details of a TAS rather than the point of the category. This makes the category non-future-proof, needing to be amended with a potentially ever-growing list of prohibited glitches. And which glitches to include in that list would be a subjective judgment. So overly specific category names do not provide the objectivity they at first seem to promise.
That said, I do think it is good to maintain a list of which glitches are being avoided for each category. One could decompose categories into a "statement of intent", describing what the category is about and allowing it to adapt to new discoveries; and "current interpretation", a list of the glitches that are currently deemed to be excluded from the category. But I think the "statement of intent" part, such as "less glitched", makes the most sense as the category name, as it is
short, descriptive and unchanging. The "current implementation" part is a detail better mentioned in the movie description.
Default constraints
For the second issue: Any%/unlabeled TASes are set of inputs that complete the game as fast as possible under the default constraints. These default constraints used to be something like
No passwords
No cheat codes (?)
Hardest difficulty unless it only adds to boss hp (?)
No game-breaking glitches
Deviations from the default constraints must be mentioned explicitly. For example, a 100% TAS has an added constraint of having to collect every item or similar, and this is mentioned explicitly in the category because it's not part of the default constraints. Similarly with glitched categories.
Recently, "no game-breaking glitches" was removed from the default constrains, turning "glitched" categories into "any%" and "any%" into "no foo-glitch" or "less glitched" or similar. But this hasn't been done consistently. For a glitched game with many categories one would typically have a single "glitched" category and many other "less glitched" categories such as "any%", "100%"*, "low%", "pacifist" etc. So to be consistent, not only "any%" should be renamed to "no foo-glitch", so should "100%" be renamed to "100%, no foo-glitch" and so on with the other less glitched categories for each game.
I think whether "no game-breaking glitches" should be part of the default constraints depends on what constraints the majority of categories use. If almost all of them assume no such glitches, then factoring that out into the default makes sense, like with the "no passwords" constraint. But I'm fine with either way, as long as it is done consistently.
*: We do have a few examples of glitched 100% runs, though, for example from the castlevania series.
I've ranted on about this issue in IRC several times before. I don't mean to present a full argument here, but rather to sum up my thoughts and have them on record somewhere. I'll try to keep it short because I know I would not read my own rambling if I wasn't me.
I don't know where and when we got the idea that "glitched" was not a valid category name. There is a bit of wiggle room regarding its exact definition, sure, but the meaning I always assigned to it was for a movie which would use one big glitch or combination of glitches to bypass a large chunk of the game's sequence in one fell swoop. A run that instead just contains glitches generally still keeps the game's sequence mostly intact and recognizable to someone who knows the game in question.
The differentiation was made for viewer benefit, because with many of these games, a TAS playing through the game normally was just as interesting as the glitched route, because of more gameplay, strategy and so on in the run. A good example of this are the GBA and DS Castlevania games: Why have a 3 to 5 minute run that warps straight to the end when you can have a 25 minute run that shows off the entire castle? Additionally, a run being "glitched" generally implies the run being the fastest category and faster than any other branch by a considerable marigin, barring things like "NG+" modes.
So with that in mind, while we don't necessarily have a concise and entirely consistent definition of what "glitched" means, I think that most of us have a very precise expectation in our head about it. I've heard the point raised by adelikat that this does not apply to newcomers, but I think that they can learn by simply watching 'glitched' runs and learning for themselves what that term means.
But what about multiple "glitched" runs of the same game, such as with Super Metroid?
The main point of a branch name is to define a movie to give a quickly recognizable outline to the audience. The word "glitched" allows us to get a very clear understanding of what the run is about, but it's not the only one. The same way, you can name specifically the glitch the makes the run what it is. In the case of Super Metroid, one run uses the X-Ray scope to tunnel into the final area early. It could hence simply be the "X-Ray glitch", people familiar with the game's TASing side would understand what it's about - those who aren't would find out.
This brings us to the other side of the table:
Ever since we had a Super Metroid "less glitched 14%" run, I knew we had a problem on our hands. The main problem, as far as I'm concerned, is that most of the movie branches for the more branch-heavy games read horrifyingly awkward now. They also don't really fit established conventions for the most part: The "11 exits run" for Super Mario World has only been the "11 exits run" since we discovered it could be completed faster using Total Control. But why is it the 11 exits run? We could just as well glitch the game into skipping the final level, making it a 10 exits run. The reason we have an "11 exits" run is that before TC, it was the established "any%" branch of SMW.
It's not defined by the fact it's 11 exits. It could as well be 12 exits if the game was programmed that way or it somehow turned out to be faster to complete an extra level. It's defined by the fact it's the fastest run you can get without glitching the game. What does that make the current "" run?
In the same vein, many of the "no X glitch" branches we have were named retroactively as a reaction to glitches that have popped up after those runs were made. It was not the point of the authors to make a run without such a glitch.
I also agree with Meshuggah's statement:
MESHUGGAH wrote:
TASes should be branched individually using the game's (speedrunning) terms and tagged with general things.
This especially applies to games with a strong speedrunning community like, you guessed it, Super Metroid. This allows us to use established terms many people are familiar with already. I think that generally, individual games' speedrunning terminology should take precedence over our naming scheme, regardless of what we end up chosing in the long run.
In the end, you have to consider that there will always be some grey area and uncertainty, as newer and crazier runs, such as the "Super Mario World TC 11-exits" run of this year's April Fools shows us. We will not be able to get an entirely consistent naming scheme out for every movie. Games are just too different from eachother for that. However, our audience aren't robots. What do we need a rigid and perfectly consistent naming scheme for if it just ends up throwing people off who are familiar with the "old way" while being stupidly awkward in some cases. What I think most people have forgotten is that we can happily allow for some inconsistency because ultimately, the runs matter, not their names.
I hope my thoughts were somewhat coherent and followable. I may end up editing this post to add and remove sections and add emphasis later on. Cheerio.
I think that the motivation between the branch naming change was that the role of branch names is not to describe a particular aspect of the run, but rather just identify the branch in an unambiguous manner. (If the branch name happens to be descriptive, all the better, but that's not its main role.)
Another motivation was that we wanted to make clearer which one of the several branches is the "official word-record TAS". This could have been a certain chosen branch name ("any%" probably being the de-facto name for such branches previously), but a consensus was reached that the only unnamed branch of the game serves this purpose well enough.
The role of describing distinguishing aspects of the run was relegated to the tags that can be applied to a publication. If, for example, a TAS abuses so many glitches that it deserves mention, then the correspondent tag is preferable to doing it in the branch name.
If the "glitched" TAS also happens to be the fastest completion of the game by any means, then that branch name would serve solely a descriptive purpose, which is really not the role of branch names. Moreover, it would be confusing because it wouldn't distinguish which one of the (possibly) several branches is the "official world record".
I'd say: Either have the official fastest completion unnamed, or give all of them the exact same branch name. Do not start giving different games different branch names for their main fastest WR completion branch, because that's only inconsistent and confusing. In the discussion that started the change it was the consensus that being unnamed is ok.
If this means that basically no run will receive a "glitched" branch name, is it really such a loss?
Btw, if a problem is that the category of a run is not clear enough, then perhaps we could come up with some kind of minor redesign that makes the tags of a run more prominent.
Perhaps there could be two types of tags: "Major" tags and "minor" tags. The major ones could even appear on the run's name line (ie. the one that appears as a link to the run's page, and which contains all the condensed info of the run in one line.) The minor ones could then just appear on the run's info box as normal.
However, we shouldn't start mixing branch names and tags.
for the sake of clarity it should be
[glitched / glitchless] [low % /any% / 100%]
the glitched any% run should be the fastest one, if a glitched run is cooler but longer , then it should become a moon candidate.
branching for a specific glitch would be too arbitrary and subjective , it would become too complicated to sort...
now the real question is "what about total control" ? should it be considered as a glitched % or as a separate category ?
if it is to be considered a gliched % then we will be seeing a lot of glitched run obseleted by total controls run, while packet injection is always coool it is kinda creating a pretty common and predictible kind of run that would always goes as "we do strange stuff, we inject code, we trigger the end" and would kill on the long term the entertainement and diversity of the glitched branch....
I completely understand and agree with having movies that show off more gameplay than the fastest run. But here's the problem: for almost any speedrun on the site, it is possible to make an alternative run that shows off more of the gameplay. However, in many cases these would simply be (1) suboptimal and (2) arbitrary. So how does one distinguish a "non glitched" run from a "suboptimal" or "arbitrary" run? We need to be able to answer that to stay true to the site goals. Even in the few hours that this thread has run, it already has half a dozen different definitions; that doesn't help.
Warp wrote:
I think that the motivation between the branch naming change was that the role of branch names is not to describe a particular aspect of the run, but rather just identify the branch in an unambiguous manner. (If the branch name happens to be descriptive, all the better, but that's not its main role.)
Another motivation was that we wanted to make clearer which one of the several branches is the "official word-record TAS". This could have been a certain chosen branch name ("any%" probably being the de-facto name for such branches previously), but a consensus was reached that the only unnamed branch of the game serves this purpose well enough.
The role of describing distinguishing aspects of the run was relegated to the tags that can be applied to a publication. If, for example, a TAS abuses so many glitches that it deserves mention, then the correspondent tag is preferable to doing it in the branch name.
If the "glitched" TAS also happens to be the fastest completion of the game by any means, then that branch name would serve solely a descriptive purpose, which is really not the role of branch names. Moreover, it would be confusing because it wouldn't distinguish which one of the (possibly) several branches is the "official world record".
I'd say: Either have the official fastest completion unnamed, or give all of them the exact same branch name. Do not start giving different games different branch names for their main fastest WR completion branch, because that's only inconsistent and confusing. In the discussion that started the change it was the consensus that being unnamed is ok.
If this means that basically no run will receive a "glitched" branch name, is it really such a loss?
Btw, if a problem is that the category of a run is not clear enough, then perhaps we could come up with some kind of minor redesign that makes the tags of a run more prominent.
Perhaps there could be two types of tags: "Major" tags and "minor" tags. The major ones could even appear on the run's name line (ie. the one that appears as a link to the run's page, and which contains all the condensed info of the run in one line.) The minor ones could then just appear on the run's info box as normal.
However, we shouldn't start mixing branch names and tags.
I endorse this post. Finally, it is worth remembering that the category "glitched" was never common (it was used by about 16 runs out of 2400 on the site), and was never consistently applied in the first place.
Currently, ""glitched" & any%" seems to be winning, so I'm wondering does that mean it'll be brought back even though noone has been able to define "game-breaking glitch"? :o
Edit:
scrimpeh wrote:
There is a bit of wiggle room regarding its exact definition, sure, but the meaning I always assigned to it was for a movie which would use one big glitch or combination of glitches to bypass a large chunk of the game's sequence in one fell swoop.
Explain why Super Mario 64 "0 stars" was never considered to be "glitched" please yet Link to the Past runs are. Are you saying skipping the 70 star requirement isn't a "large chunck" of the game?
I was always thinking of "any%" as the definition of a "anything goes" run, the absolutely fastest way to beat the game, no matter what it takes. From there on, the communities can make different variations of any% that ban certain techniques ("any%, no X"). For me, that's what any% is about.
Therefore, if the fastest way to beat the game is not called any%, this contradicts the meaning of the term. If "glitched" is faster than any%, why is the latter even called any%? It's not a true any%, it's just another type of "any%, no X", so why don't we use this naming convention instead? This way, there won't be confusion regarding the term "glitched", just different types of any% that explain their restrictions in the title.
Current project: Gex 3 any%
Paused: Gex 64 any%
There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
Why not just use "conventional route" for the conventional run? (This kinda also references the recent "No Unintended Exits" run of Zelda 2)
You mean like theseruns? Both use the "intended" route but skips every single battle within said route, while still technically going to those battles by forcing the opponent to just quit. :P
Edit: The latter's previous movie was also referred to as "glitched" despite using the "intended" route yet the former example isn't. >.>
Currently, ""glitched" & any%" seems to be winning, so I'm wondering does that mean it'll be brought back even though noone has been able to define "game-breaking glitch"? :o
Edit:
scrimpeh wrote:
There is a bit of wiggle room regarding its exact definition, sure, but the meaning I always assigned to it was for a movie which would use one big glitch or combination of glitches to bypass a large chunk of the game's sequence in one fell swoop.
Explain why Super Mario 64 "0 stars" was never considered to be "glitched" please yet Link to the Past runs are. Are you saying skipping the 70 star requirement isn't a "large chunck" of the game?
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.
Heavy glitch abuse is when one does sequence breaking by some in-game means (game engine faults). Soft sequence breaking. Glitched is when one corrupts the gameplay and breaks the logical conditions for common soft sequence breaking, overcoming the gameplay limitations. Hard sequence breaking. What do you think?
I disagree with this, since it still wouldn't really work for movies that don't have memory corruption glitches but are heavily messed up like Link to the Past (It's a simple clip glitch, which means the movie that doesn't use it would still have "no clipping/mirror-jumping/mid air-saving/variant of exploration glitch").
Currently, ""glitched" & any%" seems to be winning, so I'm wondering does that mean it'll be brought back even though noone has been able to define "game-breaking glitch"? :o
Presumably we'll vote on the definition next.
Although considering both sides of the issue are fairly large (and we have a significant amount of people voting for a third option), it would be a sensible approach to find a common middle ground.
andypanther wrote:
I was always thinking of "any%" as the definition of a "anything goes" run, the absolutely fastest way to beat the game, no matter what it takes. From there on, the communities can make different variations of any% that ban certain techniques ("any%, no X"). For me, that's what any% is about.
Therefore, if the fastest way to beat the game is not called any%, this contradicts the meaning of the term. If "glitched" is faster than any%, why is the latter even called any%? It's not a true any%, it's just another type of "any%, no X", so why don't we use this naming convention instead? This way, there won't be confusion regarding the term "glitched", just different types of any% that explain their restrictions in the title.
There is a bit of wiggle room regarding its exact definition, sure, but the meaning I always assigned to it was for a movie which would use one big glitch or combination of glitches to bypass a large chunk of the game's sequence in one fell swoop.
Explain why Super Mario 64 "0 stars" was never considered to be "glitched" please yet Link to the Past runs are. Are you saying skipping the 70 star requirement isn't a "large chunck" of the game?
The difference between those two lies in the evolution of the 0 star. It started as a 16 star, became a 1 star and finally a 0 star. The original 16 star runs was probably considered "too intact" to be a glitched run, and with lower and lower star counts obsoleting the 16 star run, the original definition still is used.
I would also say the SM64 run is not considered "glitched" because it still keeps some of the game's structure, namely doing the Bowser levels and fighting him the regular way, unlike ALttP, which just skips straight to the ending sequence three minutes in.
Additionally, the terms 0 star, 1 star, 16 star, and so on are well established in the SM64 speedrunning community, with all of them having a very specific route associated to them. This doesn't really apply to ALttP.
