We had a (very short) discussion on this very subject in TASVideos Discord just a few days ago:
For one, I'm positive that if any further improvements are to be found, that's how they will most likely be found.
Life stealing I am completely okay with. I was talking specifically about continues. Now I don't know if EZGames and I misunderstood each other, since I was under the impression that it would end up using continues to afford the constant game overs for the second player, but if only life stealing is used, I have no problems with it whatsoever.
Honestly, I wish you didn't use them at all, especially since it's not that much of a timesaver. I'm very predisposed against using continues because they are an unlimited resource box intended to enable the less-capable players to buy their way past the skill checks, not unlike e.g. level passwords or overly powerful option menus. (Extra lives are somewhat similar in principle, but unlike continues and passwords, etc., they exist within the design space of the game; that is, games are designed with a certain life stock in mind and put score- or item-based ways to increase that stock to manage difficulty.) If this were an arcade game, it'd be an automatic no from me. So in my opinion, if you decide to tap into such a powerful resource, it better be worth more than a couple percent of the total run time lest it become an entertainment liability.
Anyway, what's done is done I suppose. I won't insist you start over, just rantingexpressing my stylistic preference, kek.
Please provide the complete input. Regardless of which version will count for the timing of the run, and whether the submission is or isn't published, this is a common courtesy to the encoders who should not be expected to optimally finish the game in your stead to capture the ending in its entirety.
The native resolution of IGS PolyGame Master boards that Ketsui, Espgaluda, Dodonpachi Daioujou and some other games of the era run on is 448×224 (or 224×448 when vertically oriented), which is automatically corrected by the standard 4:3 CRT arcade monitors. Multiple other hardware boards (such as the CPS) also have non-square pixels.
I agree with adelikat: leave that for the judges to decide. Nach had that opinion in 2010, somebody might have a different one in 2020.
We don't really have a precedent right system where every executive decision made in the past holds true forever. It's much better not to discourage people from pushing the categories they like, even if they don't pass the judgment on entertainment or variety merits. Who knows, some day they might.
This, kind of.
There were also some practical experiments involving counterbalancing gravity and sun's heat energy, but that's wonky and does not work continuously for obvious reasons, let alone produce anything useful. At least the clock tells time.
Haha, that is beautiful! The translator took the word "dump" a bit too literally there. :D
If anyone here can communicate this mistake to the Citra devs, please do.
Note that the French version also doesn't differentiate between the "movie" and "video" in that menu, causing even more confusion.
"The bomb-hoarding hero is back, and now he has a partner on a suicide watch."
Just to confirm, you aren't using the level skip glitch here, right? And the reason to constantly kill off P2 is lag reduction or something?
I'd like to address this one point. This is actually not a conflict—it is a forward compliance perspective.
When making a new (i.e. first-generation) TAS, it has to do better than what humans are doing at that time. Otherwise there is no point bothering with it when you can just watch the unassisted run and get the same experience.
After the publication, it might so happen that humans find a new strategy that a TAS cannot improve upon.
Because of this, when making an improvement to an existing publication, it has to at least match what it cannot improve. Perhaps it should be clarified that directly comparable sections should be matched, not just final time (e.g. by being faster in one place but slower in another). It should not be slower anywhere, but if it cannot be faster, it doesn't have to be.
Mandating that an improvement outperforms a human run across the board would have been impractical for cases where there are no further improvements to be found with existing knowledge. Enforcing such a rule would make some improvements unpublishable and have us stuck with even worse movies.
'Twas fun, though I can't claim to comprehend every little trick you've done there. But with this kind of mobility I'd watch it even if it were only comprised of the platforming minigame.
Thanks for posting them. However, since you have already posted these improvements in the TMNT3 thread, which is the proper place to post them, please continue all discussion there.
Submission threads are only to be used for discussion of that particular submission, and duplicate posts are frowned upon in any case. Please take note.
Well, I'm not the one looking, but there has been astronomical research into finding evidence of curvature, e.g. in the form of repeating regions of the sky (in particular, CMB patterns). There were promising results, but as far as I know, nothing of the sort was confirmed in the end.
There is no definite answer to that, but astronomers have looked into it for decades and so far have seen no evidence of any boundaries or wraparounds. Not even a trace thereof. So it's either actually infinite or so big it is functionally infinite.
I made the thread title more descriptive.
Also make sure—both now and in the future—that you state the BizHawk version and the core you're using (BizHawk has two GBA cores).
Also don't mention any ROM sites (this is a very strict rule and you will not be warned again).
What about this idea, though?
This way the run gets published, kaan55 is happy, and and the branch is put to good use (at least in perspective). Since a low-glitch run would also be solo Sonic anyway, it can be easily repurposed with little further justification necessary (especially if kaan55 does it himself and/or a new S+T any% is submitted by that time). Since a low-glitch run would barely overlap with any other category, it'd have to be a separate branch anyway, so might as well start it now. You can even preemptively put it in the verdict that a low-glitch Sonic run would be able to obsolete this one.
Please do! There are many of us who appreciate competently done low-glitch runs of games that have been thoroughly broken otherwise. I promise you that both feos and myself do. I will personally come and gush all over it in the submission thread. :D
Also, while feos may propose unexpected (and occasionally disagreeable) solutions, he's always looking for input and is open to discussion. You shouldn't treat his proposals as the final say on the matter; in fact he is inviting further discussion precisely because the answer is not obvious and hence not yet decided. He's already said it before that the last thing he wants is players quitting over unnecessary decisions. So voice your disagreements, argue your case, and he will try to come up with a better decision.
It's actually explained (albeit in a roundabout fashion) right there where you saw it.
The password (BCDF) allows one to start at stage 1b, "Day of the Tentacle", the bonus stage that you can also play after stage 1 if you choose so (as the run does), hence being able to repeat the stage. This gives the following options.
1. Play the bonus stage twice (1b → 1 → 1b → 2...). This adds ten extra victims at the cost of, well, playing the bonus stage twice.
2. Play stage 1, then go to the bonus stage. (1 → 1b → 2...), which is what the TAS does. This skips the extra victims but avoids repeating content.
3. Play the bonus stage first, then skip it the second time around (1b → 1 → 2...). This is slower than the previous option because 1b has a lot more ground to cover, and you won't have the Speed Shoes for it if you start with it.
I think option #2 is the most sane choice because getting the victim count higher at the cost of using a password and playing the same stage twice is a dubious goal. The game doesn't track victim count in a percentage form, but the spirit of a 100% category is fully preserved here already as the run saves all victims in every stage the game has.
There's also the issue that this run is the most optimized any% to date, so there is definitely a merit in that, but when the regular Sonic & Tails any% is inevitably updated with the new tricks from this run, the difference between the two will probably be the least pronounced among all the categories. Tails is mainly used for the skips, with the rest of the movement being identical, so most of the run is still the same, and there's a certain redundancy to that.
So since it functionally is a lower-glitch, more-sightseeing any%, it should be published in a separate branch, but eventually replaced with a proper low-glitch, more-sightseeing any%, also played with solo Sonic.