Let's dig up a rather old topic and look at it in a different way: As documented by 2 threads in this forum, back in the day mugg was comparing Tycon's run to his own and came to the conclusion that Tycon's was recorded at a wrong framerate. I did some comparing on my own and came to the conclusion that it is mugg's run that is sped up. For the 3 segments that I compared so far, mugg's recording is running 0.8% too fast. I used the in-game timer as a marker, comparing the segments between the frame the timer switches to 399 and the timer switching to x for 1-1, 2-3 and 4-3, x being 361, 268 and 206 respectively. I compared the runs of callumbal, TheTycon, arkanias566, darbian and 3 of my own runs to that of mugg and arrived at the following results:

The empirical evidence clearly shows that mugg's run should be retimed as it doesn't seem to fit the speed of any other used platform (SGB2, GBP, BGB emulator). I believe this is a recording/encoding issue, rather than a platform issue so it should be safe to assume that the speed-up is consistent over the whole run. I will still compare more segments to make sure that this is the case. By dividing the correct consensus time for a segment by mugg's time for the same segment we get a coefficient. For the segments I timed so far, this coefficient would be ~1.008. With that we can recalculate mugg's time (12:45 x 1.008) and would arrive at a time of 12:51.7. I'd like to hear a few opinions on this matter. Cheers.

Gameboy games!

I'd like to chime in on the decimal topic. As Kirua said, milliseconds really only make sense if you have a proper recording and can do a frame count to verify. However, you can relatively reliably time a run down to the tenth of a second from any VOD. In many games with close competition that would help to break up ties. I know that some communities time RTA runs only down to full seconds, but not all of them do. Having an option to determine the number of decimals would be great. On the SML2 leaderboard I'm using the ugly workaround to display times like "26m 38s 200ms" to achieve the aforementioned resolution. Another flaw, imo, is that flat times, e.g. "26m 33s" for 26:33.0, omit the milliseconds by default. While displaying the zero decimal is technically redundant, generally omitting it still disrupts the homogeneity of the displayed times. While most of this is a cosmetic issue, there's also a practical component: from a board moderator standpoint, I want the times to be as accurate as reasonably possible (tenths are fine in the majority of cases even with frame drops and reencoding shenanigans as long as there's a timer in the recording) while still preventing people from submitting fantasy decimals, like Kirua mentioned. Having milliseconds enabled right now for the purpose to break up ties gives submitters the wrong idea that they can resolve their times down to that level from any recording. That puts the strain on moderators who have to fix those submissions and leads to misunderstandings about timing in general.

For now I've settled for this approach: http://www.speedrun.com/The_Site/thread/i8v46 It would be really nice if display options would be added to extra variables to make them more useful, though. This whole issue should not be forgotten since it concerns every game where emulator and console runs stand side by side.

Anyone? This seems like an easy fix to me. And it would help the moderation issue I mentioned in http://www.speedrun.com/The_Site/thread/ql9tb a lot already.

Hi there! As a workaround for the currently insufficient platform selection options I've added an extra variable to the leaderboards on speedrun.com/sml2. The background is that I think it should be mandatory to specify the emulator during submission if you used one - mainly to make moderation easier, but also to add valuable information. The problem about doing this with an extra variable is, well, I guess it's just not intended to be used for something like this. It has the nasty effect that the variable shows up as some kind of category in the run overview of a user. Example: It is also not possible, as far as I can see, to change the position of the extra column on the leaderboards. Having extra variables is a good workaround for current shortcomings of the site, but I understand that it probably wasn't intended for this. So here's my suggestion for increasing the usability for extra variables: Would it be possible to simply hide the extra variable from everywhere except the page of the individual submission?

And on a more general note: is a revision of how the site handles platforms on your to-do list? Right now it is not clear whether the platform variable is meant to indicate the systems with an official release of a game or platforms that speedruns are performed on. Currently, it's a weird hybrid that doesn't serve either purpose sufficiently.

I like the emulator full-window capture approach a lot. While it's still a workaround that requires proper moderation, it makes it a lot easier to sort out certain submissions. (And maybe it pushes one or another runner over the line to get a console setup for aesthetic reasons. :-p) (Again: proper moderation is not a given for many games so this approach alone would not change the status quo.)

@Lightnat0r

Thanks for sharing your thoughts! Let me repeat my suggestion to merge this topic with a revision of the platform database. Whitelisting particular emulators seems like the more efficient way than keeping track and blacklisting all the negatives and their derivatives and revisions. For GB/C games the decision about accepted emulators is basically already set in stone. Exceptions could always be made on a per game/category level (just like platforms are handled on a per game basis atm). Other platforms have established accurate and widely accepted emulators, too.

I feel like with some more brainstorming - as is happening right here - good and viable solutions can emerge from this.

Thanks Lightnator! That's what I get for copy-pasting. Now I broke my post completely. As a site admin, do you have any info on the topic(s) I brought up?

Thanks, Trollbear666, but it doesn't seem to work. Source: http://pastebin.com/76vadhTw <-- Anything wrong with it that I'm overlooking?

[quote=PresJPolk]If the moderators of a game you care about don't know what they're doing, contact them? I don't see the problem here.[/quote]

You're missing the point. This is not a per-game issue. It's not about "a game {I} care about", but any game on the site, be it a popular one with a big community or one of the dozens of cryptic ones with just one submission and one moderator (=the submitter) and no proof checking. And this is exactly why I'd like to see this as a fundamental site function instead of having to tackle each single false submission after the fact. The submission form is the checkpoint that every single submission has to pass through - so it's naturally the best place to implement the proposed feature. And it also the place where the submitter is forced to read information like "VBA is banned" because we simply can not expect that everyone regularly reads the forum posts, frequents speedrunning Twitch chats or is even part of the English-speaking (speedrunning) community at all. It is that one interface that we have. And it matters because we want this site to be a representative, complete and fair assembly of up-to-date leaderboards. Right? The whole platform feature might need a revision in general. Admins having to unlock certain platforms on a per-game-basis and platforms that are irrelevant to speedrunning being the sole default option(e.g. "GBC" for all Gameboy games) are remnants from the beginning of the site when that data field was probably not even intended to be used for verification, but just for quickly stating the original platform in the game info on the top left of the game's page. A revision is long overdue, imo.

[quote=kobepilgrim]Could there be a way to propose removing VBA from those types of lists, at least on the major ROM sites?[/quote]

No. This is solely related to speedrunning. Most people couldn't care less for a game running at 59.7275 fps instead of 60.3 fps. An emulator that gets rid of the lag that the original platform suffered from can be seen as an improvement of the playing experience for the casual retrogame fan. VBA is and remains a valid option for those. It's only when it comes to competitiveness and comparability that accuracy and fixed standards matter. And that is where this site and its own functions come into play.

[quote=PresJPolk]Maybe just adding an emulator dropdown for emulator runs site-wide, and allowing game by game auto-rejects, would be good?[/quote]

Exactly. Maybe I missed making myself clear in the opening post, but that is exactly what I was trying to suggest:

As soon as someone ticks the emulator checkbox there should be a mandatory textfield or dropdown menu from which the submitter has to specify the emulator he used for the run.

(P.S.: I thought sr.com supported BBCode?!)

Interesting suggestion, PresJPolk. After testing that, I have to say that it doesn't seem like a viable solution, at least right now. First and foremost it requires moderators who know and care about it. As pointed out earlier, many simply don't know what they are doing. Secondly, unless I'm doing something wrong, the feature of custom variables doesn't seem to be in a fully functional state. I can't seem to change the column order, i.e. where the new variable shows up. I can also not set the default view to show runs of any option of the new variable¹. Am I missing something?

¹Edit: I found a workaround for the default view: set a default option, apply it to all existing submissions, then remove the checkmark on the default option. Makes this more viable.

I stand by my point that the issue is most effectively tackled globally instead of on a per game basis.

The problem is inaccurate emulation, both in clock speed and in lag emulation (and I think load times are affected as well in some games). With this combination of inaccuracies it is impossible to "adjust" such a time to be in accordance with console runs. They're just not comparable.

(An example from just today: VBA emulates the GB game F-1 Race ~0.92 % too fast over the segments I timed.)

Hello! I think I can speak on behalf of a large part of the GB speedrun community and here is what's been bothering us for a long time now: We're in the great situation that with BGB and Gambatte there are 2 very accurate emulators when it comes to GB speedrunning. Therefore, emulator submissions are treated no different than console runs on the leaderboards for a vast majority of games (there are, however, emulation differences in some games/categories: memory-related glitches, e.g., where emulator submissions are excluded). Anyways, there would be no issue if runners would stick to accurate emulators, but sadly VBA and its derivatives are still a widespread annoyance for leaderboard moderators and, eventually, runners of GB games alike. Having comprehensive joint leaderboards with emulator and console runs side by side is a great thing. It encourages new competitive runners to give speedrunning and sharing their results a try without having to lock in and spend a triple digit amount of money on a hobby they're not really sure about yet. However, with VBA still being so widespread and not everyone aware of its inaccuracy, many leaderboards are currently filled with invalid runs, including lower ranks and top positions alike. This issue has been known for many years already and there are still VBA submissions coming in left and right. My stance is that I don't want to exclude emulator runs in general, at all, but rather spread the word about accurate emulation and help to fix the current leaderboards accordingly. Above all, leaderboards should be fair and achievements compareable. Past submissions obviously would have to be filtered "manually". That's pretty much every single game community's individual choice and it can be tedious depending on the number of submissions. However, for the present and future I'd like to request a feature that tackles the issue before such runs land on the leaderboards. There IS already the checkbox to mark a run as played on emulator. It would help a lot if the submitter had to specify the emulator they used during the submission process (not a dropdown, but a text field, I suppose, since a fixed list would be rather long and likely still not comprehensive), coupled with a note that VBA is NOT a valid option - maybe appended with a recommendation of BGB and Gambatte. If there was a way to make that mandatory for all GB games that would be ideal. This could, of course, be applied to any platform for which accurate emulators exist. I'd like to hear a few opinions on this - from speedrunners and site admins alike, preferrably.

Thanks & see you around!

Just for documentation. Mugg said someone else had reported the game freezing just as the Can Can ended.

http://pastebin.com/MiUWjStf vlackSR has more cool stuff in his channel info and on his YouTube.

The fact that I can get the jump on 2 different frames makes me care about subpixels less.

Edit: it's faster. It has a nice setup. I'll go for it in runs, eventually. ^_^

I only tried from 1 savestate, but could get it somewhat regularly after a short warmup. Needs at least a 2 frame jump from the first to second flying platform. Even got it twice in a row here.

Okay. Thanks for explaining the technical details! Since you understand those details better than I do: would you say that the current world record in Rebirth - 1 character by Zamiel is tool-assisted or not? I stated above that, depending on the game, there can be and have been "special" rulesets for many particular games like omitting parts of runs from being timed (skipping an intro cutscene, disregarding after-credits inputs, etc.; choosing start and end points of runs that are convenient to split on or retime) or starting runs from a fully unlocked save file, etc I see them as convenience adjustments for the runners, the audience and moderators who verify runs alike. It is up to each community to weigh those adjustments against the requirements of what constitutes a speedrun. In the case of this very game here, there have already been several of those "convenience adjustments" that build upon each other:

1. The game is run from a 100% save file. (Timing starts after all characters have been unlocked instead of from a clean save file like it would be required in the majority of speedruns)
2. BoI run timing starts after choosing 1 or several starting items through consecutive resetting. (In the majority of speedruns this would be included in the timing.) Don't get me wrong, I'm not arguing against those adjustments. However, adding a 3rd layer where a script(!!!) further simplifies the process of starting a run is - in my understanding of a speedrun - not acceptable. This is a no-brainer for me. A tool is very clearly (and openly) used to assist the runner. Is it a convenience adjustment? Sure it is. But does that still fit into what comprises a speedrun? Clearly not, in my opinion.

As to the missinghud2 topic, I lack insight on the actual effect of it on speedruns above a certain "skill level". I would guess that experienced runners would be somewhat aware of their stats regardless and it's mainly there for the convenience of the audience? I'm too much of a layman in Isaac speedrunning to have a clear opinion on this. In any case, I don't think it can be used as an argument to validate the use of an input script so the topic should be dealt with in its own context, IF it is even necessary to discuss it at all.

From a general speedrunning perspective - if I may be so bold and claim that perspective for myself - the discussion about the legitimacy of manipulating game files with a tool/script confuses me big time. Even just reading the game's memory for the outcome of a reset (and starting a speedrun depending on that outcome thereafter) CLEARLY would qualify as tool assistance. While speedrunning communities of particular games can, of course, agree on certain special rulesets that fit the particular game, certain very basic general speedrunning rules should always be complied with for the classic categories. Every community or individual is free to maintain separate leaderboards for categories that don't meet the criteria of speedruns for the sake of fun, competition, whatever, you name it.

(There'll probably be a proper highlight of this later) I don't know if this has been observed as extreme as this - I had not seen this before. Not sure if it could be utilized for speed somehow. The first guess is that lag may play a role here.