I still don't see a problem in continuing with the goal "obtains all free-floating rings known to be collectible in each level"... Those rings, as far as we know, with the characters being used, are not collectible.
I think if one of the goals is "minimize overworld time" since we can all agree the overworld is boring (it lacks physics, glitches, speed, and other things that contribute to entertainment), that should eliminate backtracking. The end result would probably be the same as the goal "one level entry per exit." Either way, it looks like lack of death abuse becomes a consequence. but I don't remember any deaths in the previous 100% so I'm fine with that.
Backtracking for the sake of faster total completion time, to me, sounds as dumb and about as entertaining as a sonic run where bonus effects are not ignored so he will intentionally not cross the end-sign-post until a time bonus won't slow down the score tally. I know that someone made that sonic run and it was poorly received. I'd hate for you guys to make such a long run only for people to hate on it just because you were going "all out".
TLDR: Seeing time spent doing levels "slower" replaced with time spent going around the overworld and repeatedly performing the same actions in an abusable level will make a run far less entertaining to me. I don't care if the worst offence is the special levels. The last 4 levels in the wip were not very fun... that's already 5% of the exits!!
All that glide-bouncing and glide-cancelling was pretty spiffy, but I consider it kind of cheap that bonus stages were used.
Favorites were MGZ1, MHZ1, FBZ2 (I don't remember seeing the screw elevator unloading-reloading trick before), LRZ2 (the Robotnic-TV glide-bounce damage-boost made up for the bonus ending)
Some issues I had were that the LBZ1 elevator early-exit glitch wasn't used and the extra 6 rings in MHZ2 that took forever to get with the parachute could have been replaced with a group of rings near where he got the 90th quicker, probably.
Am I the only one surprised this hadn't been thought of before (BitDW red coins)?
Lol my thoughts exactly. It seems pretty obvious when you think about it.
I guess SM64 is pretty hex-friendly so all new improvements found can be implemented without redoing everything?
I remember reading that the RNG is not reset on level entry, so things involving randomness (e.g. enemy positions) will have to be redone if something is hexed... I hope I'm misremembering though.
Felipe, if your goals are stated properly, and achieved, and the run is sufficiently different from the other published categories, as well as entertaining and optimized, then you have all the ingredients you need for the movie to be liked and published :)
DDRKhat wrote:
As for your goals I reckon the following would be adequate;
-Obtains "Perfect Bonus" in all possible levels *1
*1 - Some levels have rings in inaccessible places (Inside of objects, unreachable locations) due to old level layouts from during development. Subsequently ending up in it being impossible to achieve the "Perfect Bonus" in all levels.
I still don't think that is adequate, because then you might as well zip through ARZ2, CNZ2, etc. since the goal can't be reached in those levels. Something like "obtains all free-floating rings known to be collectible in each level" (maybe worded better? :P ) would be more clear. Then, it can be explained that this has a side effect, in acts that don't have design oversights, of giving a "PERFECT BONUS" during score tally. (If anyone ever figures out how to get some of those bad rings, their run can eventually obsolete this run without changing goal wording.)
On nicodou, the "120 star production committee" (according to google translate) is working on one and they just posted a 25-star WIP. All of it is amazing but the BoB KtQ-race, 8-red-coins, and 5-secrets stars blew me away.
He had the idea of collecting 1 emerald for each zone. I think that he chose the areas where it is easier to enter a special stage. In this way the movie would be more entertaining and a bit more diverse. What say you?
I think that idea might be fine for a movie whose goals are to get all emaralds and beating the game as quick as possible (better than getting everything in Emerald Hill Zones 1 and 2, since the fastest way through all the levels would never use super sonic. But a run whose goals are 100%, being able to use super sonic as early as Chemical Plant would be better, I think.
Oh and great job with Chemical Plant! There was one loop, early in CPZ1, that you didn't jump in but maybe it wouldn't have saved time there?
It just occurred to me, that since read only does not work in dolphin, would it be acceptable to save and restart a new movie file in order to set up the save file for my actual run?
Yes. one movie can play through, sub-optimally or however you want, so that the SRAM is in the desired state. Then you can create a new movie that starts from power-on and doesn't clear SRAM.
That was really fun to watch! It was obvious that tails helped save quite a lot of time. A few comments/questions:
1. It seems like you go out of your way to NOT break the ring TVs. Is this intentional, or would breaking them not have saved any time?
2. Is it necessary to hit every checkpoint for the Perfect Bonus? According to TSC, only individual rings are necessary.
3. Since "collecting all in-level individual rings as quickly as possible" seems to be what this TAS's goals boil down to, don't forget you can take damage to save time, unless you are also aiming for a No Damage run too.
If "full credits" is impossible without bowser being beaten, then it's probably better to go with whatever allows input to be ended sooner (whether that is bonus game thank you and creepy enemy list, or just final screen with level music). Though I would also strongly be in favor of the resulting run NOT obsoleting the current run, which should be obsoleted by the next run to get the full credits (and most likely use the goal sphere in castle 1).
So we could still have the current "any%" category and the one with incomplete credits could be "glitched" (or, since a lot of people think that terminology sucks, then "broken credits" or "null-spit version" whatever). Although then I guess the branches would merge if anyone discovers how to fuck up the RAM so bad, that complete credits play from YI3. :P
Thoughts?
I'd say if the ring is possible to get at all, like tails flying back into the action, then it should be gotten, and otherwise ignored. The submission text accompanying the TAS can just say "Every attempt was made to collect each ring in all levels, but levels X, Y, and Z have unreachable rings. Consequently, those are the only levels with no perfect bonus at the end" and no one will blame you.
Was it known that you can do a gainer (forward backflip) when hovering?
Probably not useful because to gain height rapidly, you go so far backwards that a single forward backflip won't help with ledge grabbing, and that sidehopping gives way more distance?
See slowking's reply below
[URL=http://www.mediafire.com/?ydikyk75w2gu74x]Here's a quick encode[/URL]. Not publish-worthy (since there are no subtitles, intro, or reduced bitrate for duplicate frames) but it should be good enough for people who just want to watch it because they can't get the input file to play.
will this new glitch, which allows the game to end really soon, use input controls exclusively?
As in, "only use hardware available to a real player, like controllers and the reset button?" Yes. Those are the only things recorded for a TAS to be played back.
Mothrayas, Toothache, weren't the credits of those games still semi-normal music-wise? Ah well, all I know is that this recent spurt of creativity in new ways to defeat this poor engine is incredible. It's lasted over 20 years, but it's finally destroyed :P
Anyway, [URL=http://www.mediafire.com/?434v11ncw849kt8]Here's an encode[/URL] of Masterjun's version.
if the chances of encountering this glitch are 1 in a *insert huge number here* how did Anta find the glitch?
I think he means the chances of setting [the spot in RAM that determines if credits should play] to [the value that starts the credits properly]. Notice that what あんた did was a very glitchy credits with improper background, no music, and messed up colors at the end, which I guess isn't acceptable as "beating the game". And just because you can achieve something by forcibly hacking RAM values doesn't mean that you can achieve the same thing within the limits of a game engine.
Voted yes for the following reasons:
1) entertaining to see more of the stages improved without zipping
2) innovative new strategies
3) strict adherence to stated goals, verified by 2 other members familiar with the game's physics (namely Derakon and marzojr)
To add to the discussion, I would have liked to see more of SB2 and SB3, but I also understand the necessity to not be arbitrary in goals. The aim isn't to show as much terrain as possible, so the people crying foul should also consider there are a few instances of running on top of the screen and bouncing so high on enemies that you don't see much of the level at all... If a run was made that shows those last 2 stages with stated goals like "no zipping and no ejection of more than 8 pixels in any direction", I doubt people would be appeased.
I'm going to say... no, since Masterjun loses 2 more frames than before after dying in yi1, though he gains both back during the yi4 fadeout, heh.
And he was 1 frame ahead when yi1 started so this is actually 1235 frames better than pub (I compare to when the mario sprite disappears on the overworld before the save/quit dialog: 12487 vs 13722. Please correct me if I'm doing that wrong.) :/
EDIT: I see that dp1 starts 1 frame sooner, so whoops, 1236 is correct :)
Wow, Mister! Now taking your latest yi1 (early forced camera shift seemed to help :D ), yesterday's yi4, and yi3's modified ending to keep yoshi from 2 days ago, the testruns are 1235 frames ahead of the published run. Which means that, if the rest of the run isn't slowed down by starting dp1 small, the run's maximum time would be 10 minutes and 1 frame!
Edit: The coin does fall awfully slow... I guess if you were to release the red shell a lot sooner, so it can catch the flower when it's coming out of the pipe, then jump the rest of the way, use yoshis swallow in a jump, catapult yourself off of him furing that jump, directly into that stone, that might work... or not. :D
The red shell fire is pretty slow compared to sprint speed, so hoping to get the piranha plant before it's high up would involve slowing down too much before the piranha spawns and would likely waste more time than waiting for the coin to fall slowly, unless an L camera scroll helped. ...However, that makes me wonder if we fly high enough and kill the piranha with the downward-diagonal red shell fireball, maybe that might allow for quicker coin grabbing. If it's too quick, we can always use R to force the camera to scroll for an early chuck spawn.
I'm pretty sure the optimal strat for getting the reserve goal sphere here involves either a diagonal red shell fireball, some forced camera shifting, or both.
