Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
CoolKirby wrote:
By the way, there are only two movies on this site out of almost 2500 that spell out two specific glitches avoided, so it really isn't a big problem as people make it out to be.
The point isn't really that this is a very common or big problem. The point is that it is
Not future-proof: Once new glitches are discovered that let you achieve the same thing, the categories need renaming.
Unclear to non-experts: As Potato Stomper mentioned, the categories are not as easy to interpret for casual viewers as our previous system.
Its objectivity is only skin-deep: Since the glitches to be avoided are listed out explicitly, the category names will need to be updated when new glitches are discovered, as mentioned above. This must be done by judges and the community when a TAS using a new glitch is submitted. This judgement will be just as subjective and case-by-case as the judgement whether to put something in a "glitched" or "less glitched" category.
That said, I think it is valuable to have a list of the glitches avoided in a non-glitched TAS somewhere, with the understanding that that list is necessarily incomplete and subject to future changes. But I don't think it belongs in the category name.
Here are the schemes we have discussed in my order of preference.
any%, any% less glitched, 100%, 100% less glitched, pacifist less glitched, RBO less glitched, ...
any%, any% no foo glitch, 100%, 100% no foo glitch, pacifist no foo glitch, RBO no foo glitch, ...
I included "100% glitched" in the list, but to my knowledge we have never had any runs of this category. I think the only category where we have had "glitched" TASes has been any%, which is why it makes sense to only mention it there instead of mentioning "less glitched" or "no foo glitch" in every other category. Radiant disputed this, but we did not reach an agreement because we didn't have enough detailed knowledge about the individual games we used as examples.
Perhaps we should have a poll about these alternatives at some point?
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Radiant wrote:
Spikestuff wrote:
for tasvideos there's 3.
Glitched, any% and 100%
As explained earlier in this thread, that's not the case. "Glitched" has sometimes been used here as a synonym of "any% but faster", but that's not possible since "any%" already means "the fastest run". Pirohiko is the fastest run of this game, and therefore qualifies as "any%".
Actually, faster than any% is still possible. For example by not choosing the highest difficulty (the default for any% is to use the highest difficulty unless there are compelling reasons to use another difficulty), by using passwords, or by using codes. So the definition of "any%" isn't really just "the fastest run". It is "the fastest run using [usually] the hardest difficulty, but no passwords saves or cheats". Deviating from that requires other categories, which would have to qualify for a moon under the current system, even if they are faster.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Radiant wrote:
amaurea wrote:
Regarding the "100%" category, the reason why I brought that up is because I wondered why it was called, for example "Super Metroid 100%" rather than "Super Metroid 100%, no X-ray glitch". Why is "no X-ray glitch" implicit in "100%", but not in "any%"?
The point of the movie name is not to enumerate all glitches that that movie happens to not contain. Rather, the point of the movie is to list the restrictions the run was made under (in the case of any%, that would be "no restrictions").
I agree. The only things that should be listed are things that matter for qualifying for the category.
Radiant wrote:
So the Metroid 100% run has the restriction of needing to obtain every single piece of equipment, energy tank, and missile upgrade, because that's what maxes out the in-game percentage counter. As far as we know, using the x-ray glitch does not make this run any faster, and therefore "no x-ray glitch" is not an actual restriction on the 100% run.
Here I think you are wrong. If I haven't misunderstood, there are several known improvements that can be had using the X-ray glitch. For example skipping the baby metroid cutscene (not the same as escaping from the baby metroid). The spacetime beam would also trivially speed up a 100% run, since that run already collects all the beams needed to activate the glitch (one can also use the spacetime glitch to collect >100% items, as well as lots of other fun stuff. If super metroid is going to get an EAC TAS, then my bet would be on that glitch). It is also possible to use a variant of the murder beam to pass through gray doors, which might also lead to speedups. Hopefully an expert can dismiss or confirm this, but I thought these were avoided in the 100% category because it would make it "glitched".
Radiant wrote:
Now it is possible that, at some point, someone will make a faster 100% run which uses the x-ray glitch. At that point, the most likely outcome is that this will obsolete the existing 100% run because it's faster. Only if the jury members decide that for this game (which already has seven branches) an additional branch is warranted, that this is meaningfully different as well as entertaining enough for moon tier, only then will it be necessary to use "100%, no x-ray glitch" as a branch title.
Fair enough, I'm OK with that.
By the way, you did not respond the the case of a spacetime-beam using TAS being submitted to the "no X-ray glitch" category. Just to be clear - this is an actual, not hypothetical glitch that can be performed after collecting all the beams. The fastest way would probably be to defeat phantoon (for gravity suit, to make maridia faster) and draygoon (to open the plasma beam door), and then proceed to make the planet blow up by shooting the beam in the correct room. (It would be a pretty entertaining TAS, actually.) This would have previously fallen under the "glitched" category. Should that obsolete the current no X-ray glitch TAS? (Actually, I wouldn't mind if it did, though the ideal would be to have separate categories for each of them in my opinion).
Radiant wrote:
We don't want a run for every possible permutation of glitches.
(Well, speak for yourself. I wouldn't mind that, as long as the less entertaining ones are published less visibly. If only we had some sort of lower visibility area of the site, you might call it a "vault", in which we could house these... Also, as all the SM64 youtube playarounds show, pretty odd categories often result in very fun TASes, such as the "no foo-button" or "no control stick" etc.)
Radiant wrote:
amaurea wrote:
The "no game-breaking glitches" criterion is implicit in almost all our categories, which is why it makes sense to make it explicit when it is not present, rather than the other way around.
No, it really isn't.
Are you sure? Consider games that have a category which skips most of the game with huge glitches (A), a category which doesn't do that (the "no foo-glitch" category) (B), and some other category, like 100% (C). Doesn't C usually avoid the same glitch B avoids, even if it isn't stated in the category name?
For a concrete example, consider Super Mario World, which has a 100% (96 exit) category. This category avoids the glitches that allow arbitrary code execution. If it didn't one could highjack the game's logic to give Mario huge movement speeds, or simply teleport him straight to the exit of each level. Or just do something simple like setting the "cleared" flag for each level. Once you can execute arbitrary code, the whole concept of "level", "exit" or "mario" breaks down completely, as it can all be rewritten. Do I understand the rules correctly if this 100% TAS should be obsoleted by a faster one which does this? If so, do we want the rules to be like that? (Again, not that I wouldn't want to see that TAS, but I think it should be clearly labelled.)
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Radiant wrote:
Well, the thing is that the second definition you give for any% is objective, whereas the first is not.
I agree that the second definition is more objective than the first. Not even the second is completely objective, though, as there could be games where what constitutes cheat codes is ambiguous (for examples hidden codes for activating the highest difficulty). Thankfully we have a judging system that allows us to handle these ambiguities.
I think the real issue here is how one can know whether a TAS one is making will qualify for a given category, and I also want this to be as objective and predictable as possible, just like you. I just think that "no gamebreaking glitches" is more predictable (and hence objective) than "no X-ray glitch". In practice, that is. Just consider this thought experiment. Somebody collects all the beams in Super Metroid and performs the space-time glitch to start the escape sequence, finishing the game in less time than the current "no X-ray glitch" category without beating the last boss. Objectively, this should then obsolete the current "no X-ray glitch" TAS by Kriole. After all, it doesn't use the X-ray glitch.
That isn't what you happen in practice, though. If somebody submitted that, it would be judged that the spacetime glitch is even more gamebreaking than the X-ray glitch, and the category would be updated to "no X-ray glitch or spacetime glitch" (the run would then probably be rejected for being slower than the X-ray glitch one). So the objective-sounding "no X-ray glitch" category name turns out to actually be misleading. As soon as any other game-breaking glitches are discovered, these too will be ad-hoc added to the category name. And which glitches qualify for being added this way is exactly as subjective as the "no game-breaking glitches category" was in the first place. What makes the latter better, though, is that it isn't misleading. "no X-ray glitch" would give a TASer false hope that a spacetime glitch movie would qualify, while "no game-breaking glitches" clearly states what the real objective of the category is.
Regarding the "100%" category, the reason why I brought that up is because I wondered why it was called, for example "Super Metroid 100%" rather than "Super Metroid 100%, no X-ray glitch". Why is "no X-ray glitch" implicit in "100%", but not in "any%"? And of course, the same applies to RBO, etc.. To be consistent, one choose between these two naming schemes:
any%, glitched
any%
100%
pacifist
...
and
any%
any%, no foo-glitch
100%, no foo-glitch
pacifist, no foo-glitch
..., no foo-glitch
The "no game-breaking glitches" criterion is implicit in almost all our categories, which is why it makes sense to make it explicit when it is not present, rather than the other way around.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Radiant: My point was that a naming scheme that mentions one glitch will be forced to mention more glitches if more such glitches are discovered. And what do you mean by "misunderstanding over what the term any% means"? It used to mean "as fast as possible, as long as it doesn't use passwords, game-breaking glitches, cheat codes, existing save-data or starts from a savestate unless explicitly mentioning it in the category name", and now it means "as fast as possible, as long as it doesn't use passwords, cheat codes, existing save-data or starts from a savestate unless explicitly mentioning it in the category name". Both were possible definitions, and a while back this definition was changed. That doesn't mean that the old definition was a misunderstanding, and I think it was quite sensible.
What is the current definition of "100%", and is that one due to a misunderstanding too?
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
I agree with Masterjun that categories like "no X, Y and Z-glitches" are stupid. While I see the goal of being exact and avoiding ambiguity in obsoletion chains, the effect of being overprecise is actually the opposite.
Say that you have a "glitched" movie that skips most of the game by using glitch X, and a "less glitched" movie that doesn't use game-breaking glitches to skip everything. In the name of clarity, you rename the latter "no X-glitch". But then somebody discovers that you can also skip the whole game by using a new glitch Y. Since Y isn't glitch X, this run would qualify for the "no X-glitch" category, and you would end up with two categories that both skip the whole game. To avoid that, you have to change the category every time a new game-breaking glitch is discovered. What can obsolete what is hence completely unclear from the category name, as the category name no longer describes what the category is about, which is "not skipping most of the game by using game-breaking glitches". The names "less glitched" or "no game-breaking glitches", on the other hand, are both robust and descriptive. They describe the general group of glitches that you want to avoid, rather than giving a necessarily incomplete list of individual glitches.
To be honest, though, I liked the previous state of affairs where game-breaking glitches were assumed to be absent by default, and you only used them in the "glitched" category. That is because all the other categories, such as 100%, low%, pacifist, whatever, also don't use game-breaking glitches, and hence would need to be "100% no X-Y-Z-glitch", "low% no X-Y-Z-glitch", "pacifist no X-Y-Z-glitch", etc. under the current system.
So to summarize, I think the old system made sense, was predictable and easy to maintain, while the new one is overprecise, inconsistent and obfuscates the real meaning of the categories. The individual glitches to avoid belong in the goal list of the run, not the category name.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
It would be awesome if somebody actually managed to finish the game in this category. But my main point was that one could compete in how far one could get, even if one eventually dies. That's why I suggested those category goals. They would let a TAS that dies after 4 rooms obsolete one which dies after 3 rooms, and one which kills 250 enemies in room #4 and then dies would obsolete one which dies in the same room, but only kills 220 enemies. And one which reaches stage 2 would beat them all.
I think this sounds like something that could allow for some interesting life and ammo management, herding and route choice. And since one dies pretty quickly, it would make the barrier of entry for the category relatively small. This is the sort of thing I would love to see frame wars in, much more than for SMB1, for example.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Countering sounds OK to me, and would allow the good ending. It would mostly be seen in the boss fights, I expect, since there won't be much time for countering when you're trying to keep being swarmed.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
This TAS is awesome! I'm curious about how far a no-piledriver TAS could go, though. We don't usually have categories that don't finish the game, as most games aren't impossible. But I think these rules would make for an entertaining cateogry:
No piledriving
Complete as many levels as possible as quickly as possible
Complete as many rooms as possible as quickly as possible in any incomplete levels
Kill as many enemies as possible as quickly as possible in any incomplete rooms
I think obsoletion would be unambiguous under those rules, and one would hopefully see TASes getting gradually further in the game, against those staggering odds.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Tub wrote:
a) Running at an infinite (or very high) speed, you'll sweep up an infinite (or very high) volume of rain-filled air with your front. Running slower, you'll sweep up less, which is good.
b) Even when running at high speeds, you still get the same amount of rain on your head as when standing still. Rain doesn't stop falling just because you're running! You would need relativistic effects or supersonic shockwaves for that to change.
Isn't the premise of the question that you're running to get somewhere, so that the faster you move, the shorter your exposure to the rain? Otherwise the question isn't very interesting, nor very relevant for real-world situations. The running vs. walking from A to B situation is surely what Derakon was referring to, and also what I thought of in my answer. In this case, moving infinitely fast would prevent rain from falling on your head because it wouldn't have time to do so before you arrive at your destination - no need for relativity or shockwaves here.
That doesn't mean that I support the conclusion that running is always better, though - see my previous post.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Derakon wrote:
Imagine running infinitely quickly. You'll cut a path through the raindrops in the shape of your silhouette, but no drops will fall on you. Slowing down allows drops to fall on your head in addition to hitting you horizontally, therefore it will make you wetter.
On the other hand, water that hits the body is not immediately distributed over your whole surface, so where it hits matters, not just the total amount of water encountered.
If you are only hit from above, your hair and shoulder clothes can absorb a fair amount of water, and even when saturated, some water will drip from there directly to the ground. At least for a short exposure, you can think of the upper surface of your body as an umbrella (or sponge) for the lower ones. That part will get wet, but the rest is shielded (for a while).
By running, you expose the whole front of your body (and especially the upper surface of the legs) to the rain, which could lead to a larger surface area of your body getting somewhat wet, instead of a small part getting very wet.
I think the question of whether to run or not in the rain is much more complicated that people give it credit for. The answer will depend on what sort of wetness, what clothes are worn, how long one's hair is, how heavily it is raining, the size of the drops, wind, how far one needs to go, etc.. But I think, for reasonable assumptions about the above, the answer will be "walk" for short distances and "run" for longer ones.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
I think creaothceann and Derakon bring up some good points here. If I have to choose between allowing and not allowing this, then I would prefer to not allow it. The starting RAM values should be the maximum likelihood values as determined through adelikat's suggested test.
While I agree that it techincally makes sense to treat any variable that affects sync as input in the movie file (i.e. the movie + emulator should completely specify the system), I think an arbitrary limit must be placed at some point. As Derakon mentioned, we can make exactly the same argument with regards to cosmic ray bit-flips as for initial ram state: Cosmic rays can flip any bit at any time, and arrive completely randomly (but rarely). If the emulator were to emulate this, then it would sacrifice sync-robustness. One can either arbitrarily choose a fixed cosmic ray pattern (such as no rays), or let the user specify them in the movie. But in the latter case, the TASer would be able to manipulate any ram value at will (i.e. as if he had a game genie or similar). That case is more extreme than the one we're talking about now, but they are otherwise very similar.
But in situations like these, I think the best choice isn't to choose between allowing or not allowing, but to choose both, and place them in separate categories. The normal any% category should not allow it, but the "any%, prepared ram" category would allow it. In general, I think pretty much anything should be admissible as long as it is clearly labeled.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
For any% or 100% runs, the vote is effectively about how it should be categoried: moon or vault. For any other category, however, the vote is effectively about whether to publish it or not. It is not surprising, then, that that is how people answer the poll. I.e. voting "yes" when they think the TAS is worthy of publication, even if it does not suit their particular tastes.
No matter what question one writes above the poll, it will be very hard to avoid people voting according to their understanding of the consequences of the poll. We've seen this pretty clearly through the various changes in that question through the site's history, for example in the rating behavior in the period when the poll was replaced by rating the movie.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
ais523: Yes, they're pretty similar. They both have a minor weakness which I didn't think of originally: I think it is possible to get pushed into the ground, zip along the ground, and then exit upwards (by jumping, I guess). In that case, one would be exiting from the same side one entered, and would not be taking a shortcut distance-wise. So it would be allowed. If shortcut were defined time-wise it would be disallowed, but that would be a much more vague definition.
Personally I would like to see both a glitchless and glitchy max ring TAS and I'm not sure which one I would anticipate the most. I think both should have a place on tasvideos if somebody makes them.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
S3&K doesn't even have a tangible in-game reward for collecting all the rings. I think it's questionable whether you guys would allow it, but if it was done right, and well, I don't see why not.
I think max ring runs are entertaining, and they serve to show off much more of the games than you usually see. This is what the 100% category usually does for other games. Sonic games have the "all emeralds" category (which is also fun), but unlike normal 100% categories, that one does not show off much more of the game than one usually sees. So an all rings category would fill a hole in the categories for Sonic TASes.
And I believe it's a nice solution to the aspect of S3K runs zipping through much of the core content of the game, which some people don't always like, while still being less arbitrary and difficult to define than something like a glitchless run.
Actually, I still think a glitchless run would be a good idea. And I don't think it needs to be that hard to define. How about this:
Does not take shortcuts by entering solid objects
As a part of normal gameplay the character is continuously ejected from the floor he's standing on and walls he runs into, etc. And the rule above allows that. You're allowed to enter a solid object as long as that does not form a shortcut, i.e. as long as there is a shorter path between where you enter and exit the solid area that does not go through said area.
That rule would get rid of all zipping as well as the marring clipping through the ground thing near the end of the "no zips" TAS. And I think it closely reflects what people have in mind when thinking of a glitchless run. It is also precise enough that it should be possible to apply objectively.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
The game looks really nice (except the cocky smile of the protagonist - I much prefer megaman's frown), and the transformation part looks interesting too. Let's hope the music, level design, weapons and bosses turn out well.
They should have made "free software: The game and all its assets will be released as free software under the GPL license" or similar a stretch goal. After all, the game is being paid for up front, and hence they do not depend on monopoly over distribution to make money. Just set the stretch goal high enough that they make a good profit on that alone.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
After 10 ab tests for each bitrate (which takes an eternity).
Rosenkreuzstilette Freudia stage
32: 100%, 40: 100%, 48: 100%, 56: 100%, 64: 100%, 72: 80%, 80: 80%, 96: 80%, 104: 70%, 112: 90%, 128: 80%
But it sounds fine from 72 in my opinion.
I may update with some other, less exhuastive tests later.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Those are some interesting and sensible sounding explanations for why the submission rate should go down. But as I showed just a few posts ago, there is no evidence that the submission actually is falling - it appears to have been quite stable for the last 7 years or so.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Truncated: The x-axis is months since the beginning. The y-axis is submissions per month. Time increases from left (july 2004) to right (september 2013), with one data point per month. The data set, courtesy of Ilari, is here.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
I could not distinguish reliably between 48 and orig for chrono trigger intro, last boss or lavos scream. 32, however, I could distinguish nearly 100% of the time.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
You mentioned 6, so I'll mention 6 too, in order of recollection. These are all very good in my opinion, but it is not an exhaustive list, and there are likely even better ones that I don't remember off the top of my head.
Redline
Irresponsible captain Taylor (TV)
Scrapped princess
Nausicaa (lots of other good stuff from Ghibli too)
Akira
Baccano (except the OVA)
I have not seen Shigurui or Kurozuka - I'll check them out.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
I liked the TAS, Necroyeur. It seemed very well optimized and carefully planned. I don't think the game looked that bad either, to be honest. No fast forwarding here, and I voted yes.
Experienced Forum User, Published Author, Former player
Joined: 2/19/2007
Posts: 424
Location: UK
Why not submit it? Finding out how fast it can be completed is one of the main points of this site isn't it? I think the glitchless run would be much more fun to watch, but that doesn't mean that the glitched run isn't valuable too.
Regarding the glitchless run: I think you should refrain from using stutter-walking and the menu trick, though I wouldn't have that much against the former. I would hope that a glitchless run would be able to show off some of the different kinds of enemies in the game, but I guess speed constraints will force you to manipulate the fastest enemy group at the expense of variety.
The submission rate fluctuates a lot, but overall seems to have been stable after the first 20 months or so.