Posts for moozooh


Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Fascinating. Thanks to streaming both audio and video, you get much better sound but less clean video compared to this more direct port. But doing it through controller input and accessing it through an ACE entry point makes more impressive as far as I'm concerned. It's a shame that you didn't submit it to Revision first, which was literally taking place this past weekend, and have your work join the ranks of all sorts of unlikely Bad Apple ports. In fact, one of the demos shown there on Saturday was a port of Bad Apple on the LED screens of four old Z80-based phones with caller ID. Insane stuff. :D
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
KusogeMan wrote:
Instead of arguing, I'll think of the possibilites and make an effort and create content which is passable in one rule but not the other. You'll measure the result and tell me that i was in fact wrong or right, that we are going towards a more varied publishing or that it's not better to be publishing both runs. Do you believe this to be a fair approach that doesn't offend any parties?
Sure, that sounds perfectly reasonable to me.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
DrD2k9 wrote:
No one on staff is going to complain about obtaining more publishable content for the site that meets site rules/guidelines.
Well, I'm still going to complain about the threat itself. Silly as it may be to threaten a content site with content, the tone is not acceptable.
KusogeMan wrote:
I think i might abuse this loophole in the future so if you keep the SRAM CLEAR and THE SRAM anchored movies both, prepare for trouble! I advise you to heed my warning
We've heard your takes, and took the effort to understand your position, even though you didn't make the discussion easy or pleasant for us in any way by being combative, going off on tangents, and so on. So adding this line already after a mod had to intervene to calm everyone down is very much not appreciated. If we ever find ourselves in a disagreement, approaching it diplomatically is the only acceptable way to resolve it. Twisting the staff's hand to get what you want will get you promptly ejected off the spaceship instead. So let this be the first and last time I see you do this. Please don't confuse fighting games with fighting over games. We want the former, not the latter.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
I think there is somewhat of a contradiction over the fact that we'd only accept the most optimal default character for Standard while giving a "free pass" there to an unlocked one regardless of how fast it is in comparison. I mean, what if it's the slowest character? Or at least slower than the second-best default one? (Thinking of something like Streets of Rage 3's Roo here.) Would it not be better to accept all characters for Standard? But to answer the question, if we need a generic branch name, would something like "fully unlocked" work?
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Wow... a manifesto, huh? Sounds political! Well, it is for a reason. There's an ongoing development of policies at TASVideos—both internal and public-oriented ones. We've been giving a lot of thought this year to handling moderation and community management as a whole, and given the amount of sweeping changes in what we do and how we do it, I'd like to summarize our current views, beliefs, and operating principles as well as recent changes, and run all of that by the community at large to gather feedback. This, I hope, can give us pointers to what we might be doing wrong lately, or not doing right yet, or not doing at all when we should. This is your opportunity to tell the staff if there's anything about community management that makes you uncomfortable and whether it's something that came as the result of our action, inaction, poor timing, or lack of consideration. Moderation team These wonderful individuals plus the no less wonderful Memory, Samsara, and fsvgm777. No, they don't have the role, even though they (most notably Memory) do engage in moderation activities in various capacity. Yes, I'm aware this isn't exactly well communicated on the site. Anyway, on the administrative side, Bigbass recently joined as a new moderator and I was promoted to a senior moderator to take charge of the team and all this political stuff. Samsara and dwangoAC, on the other hand, have taken a step back from active moderation to focus on the more important things. You get some, you lose some. We always welcome further additions to the moderation team, of course, and I hope this post will give you a further insight into how we operate. Policies and values Ultimately, the moderation team serves the goal to maintain TASVideos—both the forums and Discord—as a place of mutual respect and constructive, improvement-oriented communication. In 1998, Warp Records issued a compilation album titled We Are Reasonable People, and I think there's a certain kind of warm beauty in the simplicity of that phrase. I take being a reasonable person as a compliment because, above all else, it emphasizes openness and flexibility. This translates to our moderation policies. Most of what we think and believe in with regards to moderation processes is laid out in Moderator Guidelines which had an entire new section added to it just today as the result of a recent internal discussion. The article itself is intended to codify how we approach and react to any given situation right now—or at least how we think we should. Moderators are also people, and we don't always perform to our self-imposed standards—which is exactly why having more people on board is helpful in staving off operational fatigue. To further summarize what is said in the article as briefly as possible, the moderator team strives to find the right balance in their actions to ensure they handle inappropriate behavior in a way that both effectively disincentivizes it and doesn't overreach in a way that creates any collateral damage that could've been avoided. Finding that balance isn't easy, and it's an ongoing, iterative effort. So expect these community outreaches to happen somewhat regularly, perhaps once or twice a year, as policy changes accumulate. Operating principles I've referred to moderators as a team and mentioned internal discussions multiple times in this post. This shouldn't come as a surprise because virtually any decision we make is a result of having the team members chime in to provide their opinion on the matter until some form of consensus is reached. We don't discuss every technical detail—usually because it would stall the operations beyond what is necessary—but overall we almost never see situations where one of the members is acting in disagreement with the group. If we do have such disagreements, we bring them up first, and this is a cornerstone of our operations. You might've noticed that both the guidelines article and site rules are left somewhat vague on purpose in terms of matching transgressions with specific punishments: this is both to respect the phenomenology and context of each particular case and to give more room for discussion within the team and any given moderator's personal good judgment to decide the punitive measures instead of applying them as a knee-jerk reaction. We currently believe this is the best approach overall, although we do recognize it's not always the quickest or most efficient. Recent changes to discuss Here's the important part. Everyone on staff who's involved in community management, and the moderation team in particular, make policy decisions in the interest of the entire community—effectively, a group of dozen or so people decides things for thousands of you out there. But due to the nature of these discussions and the fact that they're usually connected to specific moderation cases, we can't do so publicly. It should be noted that this is entirely common and almost every community of our size or bigger does the same. But keeping decision-making entirely under wraps risks creating a disconnect between our perception of what the community would want, and what the community actually wants us to do, and we're liable not to notice this disconnect until it starts creating problems—which is, unfortunately, just as common, and happened to TASVideos in the past. Considering what I wrote earlier about openness, respect, and being reasonable, I and the others would like to minimize that risk proactively by communicating some of the changes to internal policies as they happen. Today, I have two subjects for us to discuss. 1. Whether to allow submissions coauthored by banned users. The recent round of discussion on this subject hasn't been very conclusive, so, for now, we're maintaining the status quo: submissions coauthored by banned users are allowed (regardless of whether it's a courtesy coauthorship or an active collaboration). 2. Whether to synchronize bans and other such punitive actions across both the site/forums and Discord. We had no prior stance on the matter and agreed to synchronize bans moving forward, but also to rely less on long-duration and indefinite bans if the situation doesn't call for the most drastic measures. This mainly concerns isolated (i.e. non-systematic) incidents of unacceptable behavior. If you disagree with any of these decisions, please provide your reasoning so that we could work your viewpoints and arguments into our internal discussions. This would also be a good place to discuss the content of the current revision of moderator guidelines if you find any of it lacking or disagreeable, or if you feel moderators haven't been adhering to it recently (but please be civil and respectful; alternatively, you can complain to me personally in DMs). Thank you for the time.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Mizumaririn wrote:
That is a rule question that I have not just for this contest. It is not like a TAS would require continues in order to complete a game. The rule however prevents the possibility of taking intentional deaths to save time, holding TASes of this category out of standard goal. Does 2 player arcade even also counts as more than 1 credit? If they are separate branches, I can also see the possibility of adding in players at a specific moment to save completion time. Right now, only in-game codes that add content (generally) are allowed. I don't think arcade continues ever add content because they just "continue" where you left off so that rule statement just automatically implies arcade continues are disallowed. As arcade continues and in-game codes are different concepts in video games, arcade continues should not be restricted under the rules of in-game codes. I would like to compare the arcade coin continue with amiibo for the 3DS, Wii U and Switch systems in TASing. Even though they are considered pay-to-win materials and unfair in real time speedruns, as not everyone has them, they are also just a form of input. It is the nature of the unmodified game's code to respond to these types of input, and is fair game for a TAS.
The problem with coins and continues is that they're essentially turning a limited resource given or earned within the game's own logic into a potentially unlimited resource provided outside the game's logic. Within the context of arcade games, it is the player who provides the resource in order to make up for their lack of skill, knowledge, or resource management. As long as you keep feeding the game coins, you cannot lose—with all the implications it has for the TAS goals. Note that, just because starting the game or adding a player to a multiplayer game utilizes the same mechanism, it doesn't mean they should be treated the same. Continuing implies paying up to circumvent an otherwise inevitable game over; adding a player does not. I can imagine exceptions where this does enable novel and strikingly different gameplay solutions. But by default this is hardly different from a typical infinite lives cheat (also a form of input, of course).
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Speaking of, do we mandate 1CC? The Movie Rules page is worded rather ambiguously. I'd vote for single-coin-only because coins are pay-to-win both technically and in spirit.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
To note, it's not abandoned—at least I haven't written it off my plans yet. But it's a bit of an undertaking considering most of the game needs to be reTASed from scratch, and parts of it also need to be rerouted to account for faster item pickup plans. If anyone wants to lead the effort (after DTC11 has run its course, obviously), I will join it, but I don't have it in me to do things by myself. But you should go ahead and submit what you have anyway imo.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
FatRatKnight wrote:
I have been roped into joining a team by moozooh.
Confirming. EDIT:
Denial140 wrote:
I will be joining forces with Moozooh, Invariel and FatRatKnight.
Also confirming.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Signing up and teaming up with Invariel. We'll submit the team name later.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Not even a day has passed since my own promotion, and now the one and only Bigbass joins us as a moderator! Straight from the planet of the bass, probably! Or maybe it's this bass? Who knows! This dreaded ambiguity will haunt me for the rest of my life. Bigbass has been instrumental in development of console verification and personally put some major effort into verifying TASes in the recent years, as well as being a generally nice and upstanding fellow. Big reasons to be happy to have him on board. As big as the bass. :) (We still have room for more moderators! You'll be in good company.)
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Hi. I'm moozooh, sometimes known as Vladimir Sergeyev, a technical writer/editor/translator/language specialist by trade. Most of you have likely seen me around as I've been with TASVideos for ~18 years on and off at this point, most of which I spent as a forum moderator, wiki editor, and, for the past four or so years, Discord moderator as well. Right now I'm one of the oldest still-active staff members around, probably only second to adelikat. Basically a living cryptid, borderline undead. In the light of this announcement, I'd like to share an amusing anecdote. In late 2015 I was appointed starman. Yes, that was a thing back in the day; nothing to do with the David Bowie song or the John Carpenter movie. During my short and otherwise uneventful tenure, I sent a message entitled "The TASVideos takeover plan" (in Russian) to feos, who lived an hour-something away by train at the time. That's the subject line I came up with to lure him out to have a meeting and discuss our evil plans a revision to the starring and recommendations policy and related things. It was a great and memorable meeting, by the way! I did pretty much nothing else useful as a starman. Wait, what do you mean by "that wasn't useful, either"? Fast-forward some eight years, and both feos and I are senior staff. What a... coincidence, h-huh? Certainly not in any way related to that message, trust me! But frankly, all things considered, this shouldn't come as a surprise. This community is very dear to me, and I'm thankful to the staff and other members with whom I've communicated over the years for treating me well and willing to forgive my sudden and often prolonged disappearances. No matter what happens, I always come back here, and I always receive a warm welcome. I want to share this experience with other members and give back to one of the few places on the internet that actually feels like home. Hence, as of today, I've taken on the role of a senior moderator. No, not because it's an honorary role to recognize the elderly among us. This is a new role that was branched out from activities previously performed by admins, most notably Memory. (Thank you very much for doing this until now!) For me specifically, this promotion means at least three relatively new things that will become my duties on a consistent basis.
  • I'll have much deeper involvement in matters concerning moderation policy (and by extent, advise other seniors on PR, general communication, and crisis management when necessary) and the general senior-only discussions to which I had not been privy until now.
  • I'll be managing the moderator team. This, in turn, means both that the moderators can now turn to me for advice and final say in moderation matters, and that regular users can complain to me about other moderators' conduct or decisions. And I'll be the main person hiring moderators and letting them go. I'll still be performing my usual moderator duties myself, too, of course.
  • You can yell at me directly if there's some long-standing problem with moderation and nothing is seemingly being done about it or is being done backwards, and I'll prioritize it. Keep in mind that moderators and site staff are people and they can't always operate at the best of their ability or be on the same page for every question, so not every problem can be solved quickly or efficiently. But we will try to find any improvements we can implement at the time—as is The TASVideos Way™. I will conveniently ignore the fact that breaking things is also The TASVideos Way™.
As usual, everyone is also free to DM me for any other concerns—both personal and regarding other users. I can't promise I'll be able to resolve all of them, or that it will be quick, but I will listen and provide an outside-observer take on a situation, a guidance, or an explanation on how a given situation or behavior is seen through the lens of the moderation policy. On topic of DMs and guidance, time for another anecdote: just last month a well-known TASVideos member (name undisclosed in the interest of privacy) DM'd me with a question about pooping—and I gave them an earnest answer to the best of my ability. Un-bowel-ievable, right? Can you imagine the level to which I'm willing to dirty my hands (figuratively, for now) to deal with your concerns? ...Alright, I think it's out of my system now. That fulfills the toilet humor quota. Still with me? Amazing persistence. I guess I'll take the opportunity to announce some of the near-term plans regarding moderation. You didn't think I'd take this position willy-nilly without at least some planning ahead, eh? Did you, now?
  • First—and this should hopefully be ready very very soon—we will implement ModMail in the Discord server. This should streamline the reporting process and enable the moderator team to efficiently take on the more delicate cases that require minimizing personal exposure for the reporter.
  • A report feature for the forum will also follow at some point afterwards. Crazy that we've spent all this time without one!
  • There are some things that I have in mind but will not make public until I understand how to best implement them. These mainly concern internal processes.
  • In terms of policy and response quality, I'd like to aim for incremental but noticeable improvements. Just like feos and his movie rules meta-meta-play, I'll be playing the 5D chess of moderating the moderation practice. Specifically, tightening up the bottom end (going for faster and more efficient responses, more proactive involvement when necessary) and cooling down the top end (hopefully fewer knee-jerk bans and generally more professional behavior when delivering severe punishment)..
Finally, if you've spent some time with the community, have a solid understanding of how it operates, are able to react to conflict situations and bad behavior in a calm and level-headed manner, and would like to help out with moderation, DM me on TASVideos Discord @moozooh. The more moderators we have, the easier the burden on any one in particular. Almost like a Ponzi scheme! I totally could've come up with a less unpleasant comparison! Now go register in DTC11. It's more important than reading pathetic attempts at comedy in a staff announcement thread, of all places.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
feos wrote:
Major gameplay changes coming from skipping or trivializing content are usually called a "major skip glitch", and the category that avoids it (but not other glitches) is already standard.
No, that is just a relatively small subset of the wider range of glitches to which the audience might want to see somewhat more conventional solutions. Examples of glitches that trivialize or skip content that aren't "major skips":
  • Out-of-bounds travel in some Castlevania: Aria of Sorrow runs and almost all of Harmony of Dissonance.
  • Wall zips, coordinate overflow, and other similar things in pretty much every run of Sonic 3 as well as Sonic 2 and who knows how many other Sonics at this point.
  • X-Ray glitches in Super Metroid.
  • Whatever glitchy mess some of the NES Mega Man games have become, particularly the first game.
There certainly is demand for runs that are easier to make sense of, or at least ones that respect the level geometry to some extent as opposed to just going through it in a straight line. I can see Kriole's in-bounds all-souls AoS run even has a star now upon being reinstated in a separate category.
feos wrote:
Why?
Okay, fair, maybe not most of the time. But as far as I'm aware, major skips are the most common case for glitchless/low-glitch categories in games. It's mainly the exceedingly broken games like the ones I mentioned above that warrant an extra layer of limitations that cover a lot more than just major skips.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Dacicus wrote:
Because what one person finds "stylistic" is what another finds annoying?
Assuming it did answer the question you quoted (it didn't, by the way), do we have annoyance anywhere in our judging guidelines as a motivating factor? I think there is (still) some misunderstanding here in regards to this initiative. Glitchless runs are needed not because glitches are "annoying". They are needed because unrestricted runs becoming more glitched over time is a natural course of things. A core aspect of speedrunning is finding more ways to save time, and in doing so, it is inevitable that at some point meaningful content—complex and creative solutions to gameplay problems that many people enjoy seeing—is being optimized out of the runs. Common examples include things like navigating a room full of obstacles as opposed to skipping it via OoB travel. OoB is often impressive in its own right, but once the novelty wears off you begin to understand that skillful navigation of the skipped room also was entertaining—but is no longer there due to being slower overall. In some cases (i.e. route or item set optimization) this has nothing to do with glitches, and in those cases it's the job of the Alternative class to host submissions that attempt to restore the missing content that none of the Standard categories cover. But most of the time it's down to restricting the glitch use, and that in itself is common enough to potentially warrant a Standard designation. This category isn't an outlet for making things "less annoying"; that thought doesn't even enter the equation, and it'd be a big mistake to think the staff would ever want to view things in this light. It's rude, if nothing else.
Dacicus wrote:
Back to the topic at hand, I don't see why "Glitchless" would be considered unacceptable as a goal with a separate branch if "Forgoes Major Skip Glitch" is allowed, to use the wording from Wiki: Standard.
Right, and most of the time these would be functionally identical. There is, however, a relatively tiny portion of games that are so exceedingly complex and thoroughly broken that they'd warrant both a glitchless run and a run that forgoes major skips separately. Aria of Sorrow comes to mind (boy, I sure am glad the inbounds all-souls run was restored!).
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
I'm starting to think this thread was cursed with –10 to reading comprehension. Moonwalking (assuming it doesn't save any time) is an example of a stylistic choice. We aren't talking about stylistic choices. We're talking about cases where using glitches introduces major gameplay-affecting changes. So major they make people want to see a run without them: either because too much content is skipped, or because the content that isn't skipped is trivialized to an extent that makes the category less fun. Cases where TASers optimize the fun out of the movie are common, and the purpose of this category is only relevant to such cases. The goal of this thread is to come up with some general guidelines and think in advance of any challenges so that it would be easier for us decide in the future how to approach this category to maximize its purpose and fun. Any purely cosmetic changes are completely irrelevant to this whole discussion. Besides, why would a movie with different stylistic choices but the same final time be published in a separate category? That doesn't make any sense.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
feos wrote:
But what are potential problems for games where we may try to invent this category from scratch? Do we even need any kind of future proof terms for this?
The main potential problem is mismatched rulesets (which are the definitions for our purpose) between ourselves and the RTA/unassisted community, i.e. what we want to see as the optimal ruleset for the glitchless category vs. what the playerbase and the viewerbase of the game in question wants to see. In cases where there's either no community around the game or at least no universally accepted standard for glitchless runs, and we have to come up with one, there will be situations where others don't agree. In which case we have a choice to either adapt the definition for the game in question or stand by ours if we have reasons to believe it suits a TAS better. I expect this to be decided on a case-by-case basis. The obvious thing to do to minimize the rate and severity of mismatches is to exclude everything that people most certainly wouldn't want in a glitchless category:
  • Any and all forms of save/RAM corruption, including overflow glitches.
  • Level skips that don't involve any valid in-game skip mechanic.
  • Use of any leftover debug features or official cheat codes.
  • Out-of-bounds travel for cases where the game character violates solid room/level boundaries (i.e. by clipping or otherwise glitching through them). Situations where the character goes around the boundaries (i.e. by jumping over them, like you do in SMB1 to get to the warp zone in 1-2) would typically be considered fair game, but if it breaks the gameplay too much (as in, say, Serious Sam where you can skip about 1/3 of the game by doing so), it may need to be considered depending on the case.
Those are the hard rules; the things we can easily discard outright. The rest (the "soft" rules) should depend on the game and the impact the glitches have imo. I think generally makes more sense to err on the side of excluding everything that looks sus, but, for instance, I don't think we'll ever need to discuss whether to use mockball in "glitchless" runs of Super Metroid because over the years it has universally been normalized as a movement mechanic as opposed to a glitch. The operative word here is "normalized", as it was a result of decades of community consensus, but in our case we may have to preemptively normalize things based on the gameplay impact as said before.
DrD2k9 wrote:
Consider that some people feel that actions like wall jumping and clipping through walls (as in SMB1 & SMB3) should be considered glitches as opposed to simply being considered optimized use of normal game mechanics.
If we allow ourselves to get hung up on the semantics, the next thing to discuss is whether stabbing a person should be considered murder as opposed to an optimized use of the knife. Your Honor, I only used the knife for its intended purpose! You can always arrive at unintended results even if you limit yourself purely to intended mechanics. Hence, rather than the mechanics, you need to draw the line at the results you want to achieve using them, or at least follow the lines drawn by somebody with authority to do so. This is exactly why feos is talking about alleviating player/audience concerns as opposed to coming up with rigid definitions. If there's a vocal support for a run that omits something that is far too effective—whether a programming oversight or a design oversight—then it's a "glitch". Let's just go with that to make things simpler for ourselves. We're ultimately a content platform for serving runs audience wants to see. It may result in things becoming vague every now and then, but as far as I'm concerned it's better to present two run categories that the audience wants to see the most as opposed to two run categories that follow the definitions most neatly. Definitions are a convenience feature, not a goal unto themselves.
EZGames69 wrote:
My only issue with making it standard is not every game has their entire speedrun changed just because they don’t go for glitches. If a speedrun for a game happens to be purely glitchless, but has one glitch that saves like 2 seconds (in say a 20 minute run), then would it make sense to have both a normal TAS and a glitchless TAS where the only difference is that one rather insignificant change?
I don't know if it's reasonable to expect anybody to do a new 20-minute TAS for an insignificant change. Sure sounds like too much effort. In any case I don't see the issue. People (including myself at the time) were constantly fearmongering about site bloat when the site had less than 300 runs total. Now it has over 3000, a full order of magnitude more than back then, and exactly nothing bad happened as the result; in fact, it's been something to celebrate. Gee, the 2007 me must feel like an idiot (and would be right to). If the argument requires you to constantly shift the goalposts of the bad thing you're expecting to happen, it's a horrible argument.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
What Mike said is correct; the site has been moving toward accepting more content, and this includes having more categories with restrictions and rulesets popularly accepted in the relevant communities. So if the dominant ruleset for TAS stuff in the Elma community is "no divide-by-zero error", then it shouldn't be a problem to have that as a category here. The only think I'm not certain about is running individual and custom levels because the site isn't designed for that at the moment, so you can only potentially submit a full game run. But that would already be cool enough. Combining the Elma community tools with our tools could break the game wide open. Piano roll input editor in particular would be a complete game-changer.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
bene wrote:
Two of the best players in the world get together to try out new tooling and at the same time achieve something impossible.
Can you give a breakdown of the tooling used these days? I think it shouldn't be a big problem in principle to have Elma TASes accepted here.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Not often do I see TAS footage that is literally breathtaking. Those precise rotations and brake manipulations to pick up speed on a dime are nuts.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
bene wrote:
moozooh wrote:
Also, a bit of a tangent, but are there any plans for a new Elma Done Quick?
A couple of years later but finally there are news about this. After more than a decade since the last video it's happening again. The world premiere is planned for 15th July at 10pm Finnish time.
Omg, that's awesome! Can't wait.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Oh yes, I am familiar with your work on PoP. Didn't know you've made that into a (more) generalized solution since, that's very cool. I realize it's a horrible workload for a GPU, which, I'm sure, is why nobody has ever seriously considered it, but I'd be interested to know where the limits of this approach are because GPUs allow for just so much better (and cheaper) scaling that it can overcome the lack of speed just in volume. Because, say, if this is just 1/10 the speed of a consumer CPU, we're still more than good to go. If it's somewhere between 1/10 and 1/20, we're still good to go at a big enough scale. Between 1/20 and 1/50 could be problematic, at least in the near term. Slower than 1/50 is where I'd say it's completely infeasible.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
So here's a horrible idea. Bear with me for a bit. Right now, emulator botting is fundamentally a single-threaded endeavor, and CPUs haven't undergone a drastic progress in that department lately (the best consumer CPU you'll find today probably won't even be 1.5 times faster on a typical ST task than my 5.5 years old i3-8350K from 2017 overclocked to 5.1 GHz). So if we want it to handle a complex enough task, we either have to run multiple instances of the emulator and make each instance search through a dedicated chunk of input permutations, or, in some very specific cases, we isolate the game logic entirely and then turn it into a parallelized code to be executed completely separately from the emulator. Both are very labor-intensive (especially the latter, which also requires reverse-engineering and knowledge of a high-performance programming language), don't have a generalized solution, and in the former case still don't run very fast, either. The fact that emulator code is mostly not parallelizable has been a serious problem for botting and the main argument against porting it onto CUDA or another platform using compute shaders to take advantage of GPUs' massive parallelism. However, that argument has historically been built around the speed of a single emulator instance, which would most likely be slower in this case, and it's a valid argument if we want to prioritize that. But do we? What if we use it to run multiple cores, each within its own waterboxed instance, all in parallel? Sure, each one would be slower, but on the other hand, we can run as many as we can fit into VRAM without much (if any) overhead, unlike running them on a >8-core CPU. Assuming a single core instance for a 8/16-bit platform takes roughly 200 MB on average (I just pulled that number out of my ass, don't judge me), and we're using a graphics card with 12 GB of VRAM (a GTX 1080 Ti, RTX 2060-12, RTX 3060, or RX 6700 XT, most of which can be found for a couple hundred USD on the aftermarket), we can fit up to ~60 instances of our core at the same time, with a single interface for managing their inputs. So even if it makes a single core run about twice as slow as it would have on a modern 5 GHz CPU core (out of my ass again), that's still a whopping 30x net speedup for the purposes of botting in particular. And a graphics card with 16, 20, or 24 GB would result in a proportionately larger speedup still, which would make it a really damn good generalized solution in the longer term as VRAM sizes keep increasing at a faster rate than the number of high-performance CPU cores. And then there's the possibility of running 3-4 GPUs on the same machine, and you can see how well it scales in principle. We're looking at overall speedups of at least two orders of magnitude in the near-term: up to 500x could already be achievable with today's technology if my napkin math here is anywhere close to realistic. And from there, we only need a relatively minor step towards a folding@home-style parallel computing network run by other TASVideos members on their own GPUs, each able to pick the games to which they'd like to dedicate their compute resources. I've avoided the obvious elephant in the room, which is whether that's feasible to implement at all, to which my answer is: honestly, I don't know. But if it is, it's something worth considering, as the optimization problems we're encountering become progressively more complex with both old and new games, so we'll be relying on bots more, not less, over time.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Lol, hi Pearl. :D Sick run, loved it! I thought for a bit for some potential avenues for improvement, but it doesn't look like any of them compare to simply grazing more (or, to put it more precisely, minimizing the time spent not grazing). So if any improvement comes along, it'd probably come up with more efficient grazing routes that minimize idle time between bullets. Also, am I to assume you'll be also doing RF2 and/or Jet at some point? I'd love to see either, but particularly Jet, of course.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
I also got hit with an infinite redirect but it worked when I switched off my VPN. Here's a mirror for the image just in case.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
Well, I enjoyed it for what it was, regardless of the previous one. Thanks for making me howl at moments of genuine brilliance.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.