Posts for moozooh

Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Except it's a different TAS.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Good improvement.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Would this potentially enable us to TAS the Linux versions/ports of PC games that Hourglass can't presently handle?
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
What I ended up watching this season. Ongoing Re:Zero is gradually becoming better, though it has already started on a high note. Mild harem notions aside, it's one of the few successful attempts to deconstruct the premise of an everyday person ending up in a fantasy-themed world. And it doesn't pull its punches when showing the less desirable traits of the protagonist! Not on NGE's level of psyche exploration, of course, but good enough to be believable. Mob Psycho 100. Clever and insightful satire from ONE, as expected, and amazing animation from Studio Bones, even better than expected! Well done all around. Amaama to Inazuma / Sweetness & Lightning. It's much like Usagi Drop, except with blood-related father and daughter, and focused on... making food. Likable characters and their very natural interaction makes it one of the better SoL shows I've seen recently. 91 Days. A gritty Prohibition-era mafia drama from the studio that made Baccano and Durarara. Seems well-written and not at all insulting to the intelligence. I expect a nice plot twist coming up soon, so if any of you is going to watch this, better start now lest you get spoiled. Shokugeki no Souma S2. It's exact same as S1, except with no characterization and more monster-of-the-week approach. If you like your cooking showdowns, you're going to have a blast, but if you're looking for more substance and less shounen porn, you probably won't find that here. I wouldn't bother, had I not liked the previous season. Orange. Very slow and girly. I'm not sure if I like it, but there's nothing really bad about it except the glacial pace. Onara Gorou. A brilliant short-form series straight from the bowels of Takashi Taniguchi, the man behind such acclaimed masterpieces as Mr Ando of the Woods, Salaryman Man, and, of course, Mukidashi no Mitsuko. Bananya. Another short-form about a cat that lives in a banana peel, and his friends, who are also cats living inside banana peels. It's exactly what you'd expect from a description like that. Berserk. A very, very poorly made adaptation of a best-selling manga that pales in comparison even to one made 19 years ago. I'm not even sure how you can fail so badly. But since it's the only adaptation of my favorite manga, I still have to witness it destroyed by the incompetent hands of whoever comprises the production committee. There's no excuse that Mob Psycho 100, a manga that barely anyone outside Japan knows, receives incredibly well-done animation by one of the premier studios, yet a worldwide phenomenon Berserk only gets low-budget 3D with unremarkable music, laughable sound effects, and... censored nipples (yes, in an R-rated series where 90% of the cast eventually gets tortured and/or brutally killed). Older stuff Monogatari series. I've watched all of it up to and including Kizumonogatari movie 1. Loved it! Great humor, very confident sense of style to the point where it becomes the substance, infinitely lovable and well-written characters, entertaining dialogue, extreme fanservice executed in such a deliberate and confident way that you can only applaud and say, "fucking this, THIS is how you do it". Kizu is more subdued in the latter regard, clearly more suited for cinematic experience. Can't wait for the other two parts and the final TV season. Umineko no Naku Koro ni / When the Seagulls Cry. I'm about halfway through, and I still don't get what some people like about this show. In some rare moments it appears like an intelligent mystery series, but it takes only a few minutes to wash that impression away with how annoying and unnatural everyone and everything is.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
I expected Nintendo to eventually act on this like they have on ROM sites and whatnot, but this was even quicker than I imagined. It's shameful, really. They haven't done anything decent with the franchise for our money in over ten years, and they prevent its fans from doing anything decent with it for free just as well.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
If it were up to me, and me alone, I would use the "heavy glitch abuse" tag as an umbrella term for RAM and save data corruption, ACE, generic hardware glitches like the DPCM, and other instances of (near-)total control originating from glitch abuse. And then I'd just make all the runs made that way completely separate from the rest of the runs so that the two groups never overlap in obsoletion chains. :p
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
feos wrote:
I can put it another way. There's a difference between zero and null. Zero is a valid number, null is absence of any valid entity. So, regarding completion, we have any% (fastest completion), 100% (full completion), 0% (lowest completion), but do we want to have null% (completion is for pussies) competing with those?
The problem is clearly and unambiguously defining the difference between any% and null%. — We can't go by run length (absolute, relative, doesn't matter) because it would be an arbitrary variable which will bring more problems than it will solve. — We probably can't go by glitch type or method, because in some cases the glitch in question is only triggerable 20-30 minutes into the run, and in others right at the start screen. This is arguable, though. — We can't go by the amount of skipped content because 1) every speedrun skips some; 2) it's impossible to gauge precisely; 3) it's impossible to draw the line most people would agree with. It might be that, regardless of the accepted publication/categorization paradigm, it will have to be decided manually by the judges, if only because every other mechanism is even less reliable.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Replay is alright imo. Many games with built-in input recording refer to the resulting files as replay files (Touhou Project engine games' .rpy files, for instance). The reason why I like it less is that it feels less specific to emulators and shares more context with videos and the act of playing the game again (that is, replaying it). It would likely be my second choice.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Well... you can talk candidly without being rude, right? I mean I value your honest opinions and would love to read them as well, and I'm pretty sure this is exactly what forums are for. This is a subject that potentially concerns all community members, and clearly, closed discussion is what it needs the least. Though it's your call, of course.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Wait... Why can't you talk more candidly here then? lol
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Scumtron wrote:
moozooh: MESHUGGAH is continuing work on a pacifist run, not an improvement to the published run. Personally, I wasn't sold on a pacifist run being submission-worthy until feos made some test runs and made clear that, done right, it would take full application of every little thing known about the game—including some stuff that just isn't relevant when you can kill everything. As for any%, thanks to a couple recent finds from MESHUGGAH and feos, I've resumed my sporadic efforts of improving it and it's currently 65 frames faster (should be a few more by later today).
Ohh, I see now, thanks. Still, with all that knowledge, saving just over a second over a ten year old run for what I understand to be around 1/3 into the game signifies just how good the previous one was. Mid-2006 was just about when memory watch was only becoming popular here. Anyways, good luck to both of you! I love this game and eagerly anticipate the upcoming runs.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Warp wrote:
Hmm... Please define "publication". I'm not entirely sure what exactly it is, in this context.
xxxxM.html and the videos located therein. After all they're presented a lot like actual movies, with video files, "box art", list of authors, description, subtitles, sometimes even a commentary, and whatnot. An input file is an input file, however. Right now the term "movie" refers both to the input file and the publication, which is supremely confusing because they're completely different things.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Invariel wrote:
So... What is wrong with moozooh's really well defined ... uhh ... definition?
Thanks, but I can even name one huge thing that is "wrong" with it myself: it doesn't involve the appreciation of gameplay that, while not crucial to labeling something a speedrun, is nevertheless crucial to deriving entertainment from it. That's where the complications arise. The entire history of speedrun communities is the history of compromises between that which is objectively quickest and that which is best subjectively appreciated, which is not at all always the same thing (especially when pushed to extremes where entire games are skipped). There will inevitably be grey areas involved that have no clear resolution. That being said, I believe this to be the exact reason Warp has started the discussion in the first place: the gameplay appreciation bit that is lacking from any clear definition adopted so far, yet the one causing the most concerns with obsoletion chains, judgment decisions, categorization, and so on. Perhaps we shouldn't touch the definition of speedrun per se but rather establish a clearer understanding of what the community and the site's preferences are with regards to them. In other words, what are the speedruns (or their goals, etc.) we appreciate the most ("we absolutely want more of such and such"), and what are the speedruns/goals we appreciate less ("such and such will do, but not at the expense of the rest"). Because changes in that understanding influence a LOT with regards to the content policy of the site and the work of its contributors, as has been seen over the years with "boring" speedruns and the changes that the introduction of the Vault, etc., has brought. The other question is, of course, how would we do it without involving personal bias too strongly. Reconciling fans of moderate gameplay breakage with ACE/corruption fans is not a simple task, and it's clear their mindsets are very different. Not unlike the difference between mathematicians/theoretical physicists vs. inventors/engineers. One is chiefly focused on the fundamentals and theoretical limits, the other on practical implementation and its aesthetics. And there's clearly not enough unbiased statistics (eg. a statistically significant amount of well-understood entertainment rating scores from a large set of mostly the same people) to draw any automatic conclusions. I wouldn't be surprised if this discussion with all of its potential importance will end up hanging for several more years while the will and the resources for radical changes accumulate (or are drained, in the less fortunate case).
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Wow. I knew the TASes of this game were extremely well done back at the time of their submission, but to think it would be this difficult to improve upon a ten year old run by single frames is just crazy. :D (Btw, good to know you're still posting here, MESHUGGAH.)
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
FWIW, regardless of how we call the actual input/replay/whatever files, we can still refer to our publications as movies (which is arguably a more suitable context for the word, considering how the publications are presented).
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
I typically refer to it as input file or key input file in other communities as well. If at some point the site administration decides to make a move on this matter, I would support it and suggest "input file" as, at the same time, the most descriptive and the least controversial name.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
This has prompted me to come up with a comprehensive, for-the-record definition to refer to in later arguments (potentially). A speedrun is a deliberate man-made playthrough of a given game from a defined start state to a defined end state in the quickest manner, delivered by the means of providing input via the accepted means, intended to entertain an audience or provide a technical reference, that satisfies the following criteria. 1. The start and the end states are consensually determined by the community (or communities) involved to be sufficiently unambiguous and replicable between competing community members. A community may introduce its specific restrictions but should still keep the rules clear and ground level for its participants. 2. The acceptable means of providing input are determined the same way as in #1. 3a. If a game can have multiple different combinations of start/end states (such as those including secondary goals or trade-offs) that the respective community appreciates and wishes to compete in (100% item collection, etc.), internally consistent categories are created. The default category is "get from the very beginning of the game (i.e. starting from clean slate) to the ending screen", as it is most generic and applicable to the largest set of games. 3b. Any minor goals or intermittent choices that prevent the quickest end state criteria satisfaction are to be forgone to at least the extent consensually deemed to be acceptable by the community members. 4. Any tricks, techniques, and strategies, as well as preparation work, are allowed except those expressly forbidden by the consensus of community members for the purposes of entertainment, longevity, and principles of fair play. 5. The success of reaching the end state as defined by the community consensus is to be determined by the state of the relevant memory address(es) or, in case that isn't feasible, inspection of the visual output and expected reaction to providing further input. 6a. If a speedrun is designed to be done by human skills in real time, the use of automation and other aides is to be reduced to the minimum consensually established by the community members. Other rules may be put in place to penalize the reliance on aides (such as segment break penalties) as deemed necessary by the respective community. 6b. If a speedrun is designed to be tool-assisted, it must have some kind of human factor involved (such as in planning, scripting, other sorts of decision-making). Corollaries: — An accidentally quick playthrough is not a speedrun. — A playthrough without established and/or relatable goals is not a speedrun. — A playtrough that repeatedly forgoes the quickest means of attaining its primary goal past the acceptable extent is not a speedrun. — A playthrough that finishes the game in any way without completing its goal is not a speedrun (it could be considered a failed speedrun or a demonstration). — A playthrough made using forbidden means is not a speedrun (it could be considered a cheated speedrun). — A tool-assisted playthrough done without any human factor involved (aside from providing the description of the start/end states) is a simulation. Human factor—or rather the continued struggle against its follies—is the key to entertainment. All this being said, just like with judicial systems, there is the letter of the law—the rigid definition that serves as a backbone for justice to be determined and served—and there's the spirit of the law that is the common understanding of what the law represents behind the technicalities. For me, a mistake-ridden playthrough that uses entertaining tricks and glitches and tries to be efficient and innovative even when it fails to be the quickest is still more of a speedrun than something that skips the entire game.
arandomgameTASer wrote:
I mean, shoot, remember when ACE runs appeared once a week? Hardly ruined the site then.
To be fair, the reason it didn't ruin the site was exactly that it didn't last long enough to cause irreversible harm.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Looks sweet. I'd probably wait until a couple more inevitable patches before starting any proper TASing efforts though.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Dang, you're right. I was looking at the Nico upload date. This is the actual chronology: userfiles movie uploaded at: 2016-06-15 03:12:55; submission #5145 submitted at: 2016-06-16 17:43:23; NicoNico video uploaded at: 2016-06-17 23:20. Oh well. I guess that point is moot then! Edit: ...or maybe not. The userfile was uploaded hidden. By the time K&K submitted their movie, the Japanese team's TAS wasn't public knowledge, and K&K likely didn't know about it prior to submission, either.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
mtvf1 wrote:
In fact, tasvideos has accepted a slower run once. [2276] NES Super C "pacifist" by Heidman in 13:21.40 #3762: Soig's NES Super C "pacifist" in 13:00.20
In case with Super C, the slower run was also in a slower category that was created primarily for the purpose of entertainment, which allowed some leniency in judgment, whereas both the FF6 movies in question strive strictly towards quicker completion at all costs. That being said, I would have preferred this movie to be published and remain on the site until an improvement is submitted. There is merit to "it doesn't break all the known records" argument, but 1) it's still faster than what we have or might have in foreseeable future, 2) the route it takes can't be faulted for lack of optimization—it's still a properly done TAS. Edit: To elaborate. There have been a few extremely controversial submissions, namely Zelda OoT runs by P.DOT and JNX in 2007-2008 who beat then-published TASVideos records but failed to be fastest known per se. There was a compelling argument presented that the old run was extremely outdated and "having less OoT on the site is also a good thing" (tongue-in-cheek, but hard to disagree with!), but it ultimately didn't happen because those movies were much, much sloppier than GuanoBowl's old run. Ultimately we had to wait for a submission that was at least comparable in terms of optimization, but still faster than realtime and much, much faster in terms of route and strategy. This FF6 submission is at least as good as the current published movie: it is faster, it is more entertaining (which everyone has agreed upon), and the only known way to beat it is by using another TAS-only strategy, so it has none of the controversy associated with earlier dramas. Literally the only problem with it is that a faster strategy was discovered while the work on K&K's movie was already in-progress. And when their movie was submitted, it did beat all the records existing at the time of submission. If that doesn't deserve some slack I don't know what does.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Ready Steady Yeti wrote:
On another note, I feel like people often stereotype participants of speedrunning communities, especially this one, to be extremely antisocial.
The stereotype stems from the apparent necessity to devote a lot of one's free time to the asocial activity that is speedrunning a game. This often indicates that the person involved either doesn't have or doesn't want social contacts to take place during this free time due to some personal real-life circumstances. And if you read some of the personal stories posted here, you'll find a degree of truth to it, so the perception is not completely wrong nor unwarranted, even if unnecessarily broad. After all, people rarely escape to videogames or other fictional worlds out of feeling of accomplishment—the opposite is way more likely to be true.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Warp wrote:
VR could have literally thousands of triple-A games right now. The OR development kit has been around for about three years. They had plenty of time to add quick support to most existing games.
...Most existing games that people already own and are extremely unlikely to pay the second time for. I understand the sentiment (I'm giddy enough that some games such as System Shock 2 receive fan-sourced bufixes and facelifts), but hoping for something like this to happen officially is just wishful thinking, I'm afraid. Let's face it. Adding support for features that do not exist in the game costs money. Fixing bugs and compatibility problems, as well as listening and responding to actual complaints, costs money. Re-advertising the addition of new features costs... you guessed it, money. As does the redistribution, as it has never really been free for anyone. So where do you expect this money to come from? People are hard enough to convince paying for something the first time; the percentage of the original audience who would be willing to pay again would never cover the costs. What about possible reputation risks, arising from disappointingly implemented VR in games that weren't conceived with it in mind, that might affect future purchases? Publishers know this, so they won't bother. Profit margins for new games are higher—that's just the reality of things.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Warp wrote:
Because that's exactly what I wanted: To be able to experience my all-time favorite games in VR. Instead, they are dissing traditional games completely, saying that VR "doesn't work" with them, and they have zero intention of adding VR support to them.
Well, of course they don't. Catering to nostalgia in this way isn't worth the investment. Which is understandable, because in cases with many (most?) older games adding VR support entails rewriting the entire rendering pipeline from the ground up. It's more profitable to make a new game.
Warp wrote:
They are instead adamant in creating custom VR-only games that work differently from traditional games.
Do they, though? What VR changes right now is making the camera control more awkward by making it equivalent to your head turning synchronously with your eyeballs, which is not how human eyesight works (and is hence nauseating for sensitive people). But the core gameplay hasn't changed, and the selection of genres hasn't changed. Besides, if a game is neither fun nor satisfying regardless of VR support, that's really the game's fault, not the VR's. Which is, again, understandable: if you take the time to develop a future classic, you will miss the fad cash-in. Publishers don't want that. Meh, just wait it out. The initial fad will die, people will start thinking on the ways to bring out its potential, and we'll get some good games that will hopefully encourage further advancement of the technology.
Post subject: Perhaps the best track Autechre have created since 2002?
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Ok, let me explain. They may say they're convinced it doesn't work with "traditional" games, but that's still what they're doing: same old "traditional" games with a two-axis viewpoint control. I haven't seen even a single fully developed, non-proof-of-concept game that would rely on VR instead of VR being forced on it. I mean I've just googled "best VR games" and this was the first hit. Inside is the list of same fucking old games I played a decade ago—except they're now completely smothered with shadurrs and have... yes, a two-axis stereoscopic viewpoint control system strapped to your head! But nothing else changed. No new ways to interact with in-game environments, no new interface-driven narrative approaches, none of that. The original Penumbra tech demo probably did more for immersion and the concept of "reality" than all of that list combined, if only because it actually offered entirely new ways of interaction with a 3D game environment that affected the gameplay in a major way, while still using the existing technology. It used the latent potential of that technology that everyone else ignored, using realistic physics emulation to inform a player's decisions and provide them with methods of dealing with in-game problems, real-time lighting as a gameplay element rather than a graphics showcase, etc. Valve does have a point in the sense that simply putting two eye-tracking displays on your head does not a virtual reality make. VR headset is really just a glorified way to trick your field of view in a way that erases the rest of the room from it (Google Cardboard is perhaps the most convincing illustration of this). It has a very long road ahead of it before it could seriously be called virtual reality. It hasn't failed yet because it hasn't really appeared yet—what you see is the embryo of the technology to come, which companies are eager to turn into buzzword and cash in before it gets good and they can cash in again. I mean when you saw Tron or Money for Nothing music video in the 80s you didn't think CGI "already failed", right? It barely even existed back then.