Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!
You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!
It appears that you are using ad block :'(
Hey, we get it. However this website is run by and for the community... and it needs ads in order to keep running.
Please disable your adblock on Smashboards, or go premium to hide all advertisements and this notice. Alternatively, this ad may have just failed to load. Woops!
1) It's something to give additional utility to a single din's that's a short enough commitment to use against characters she has the most difficulty in neutral against without requiring her to have already won neutral in the first place in order to safely place it (those same characters were also often the best at preventing a 2nd/3rd from being placed as well as safely dealing with existing ones!). To get an idea of the speed improvement, it was hitbox as early as 46 and 59 total (ground only. ending in the air was something even higher) before, and in 3.5 is 33 earliest hitbox and 46 total. More rounded effectiveness among the cast along with new tricks to explore basically.
Link's return boomerang is disarmed on shield though. Once shielded it can't hit you again (though you need to go out of your way to try to catch up to it), same with din's except you can refresh it by reactivating it. Is it like actually disappearing entirely when blocked? If so that's not supposed to be happening. The bizarre interaction specific to shields compared to how hit/clank worked before 3.5 was neither intended nor liked by much of anyone on the DT and only going to stick around as long as we didn't know how to fix it btw, so that was less of a 3.5 design choice and more an inevitability.
2) The higher initial speed helps to more quickly place them at a distance, complementing the return functionality as well as more opportunities to be used to continue pressure after things. It also reflects the return speed, which has the overall more useful slower movement by charging it. That said, I think I might agree that the minimum distance you can place them may be a bit too far from her to help as intended against good dashdances since I think if fast characters move in quickly enough when they see you cast it ends up a bit behind them even if retreated, and while placing it may now be safeish to do and not directly punishable there's very little if any time to act before they reach you, which includes calling it back in to help the tight spot you're now in. If it could be placed just a bit closer without affecting the move elsewhere so they can't ignore it on the way out in that situation you'd get more out of giving up that stage position in order to place it.
You can see what I mean here where CF is at DDish range and moves in 1/4 second into the side-b (she flashes red on startup making it pretty distinct, though it doesn't show up with hitboxes on). Compared it to 3.02 as well, and was surprised at how awful it apparently was at preventing them from just running in despite being able to place them so close because of how long it takes to get the hitbox out. Hard to actually see the hitbox since it's so tiny compared to the GFX, but it doesn't start hitting until that 2nd pause when CF is already next to her. Kind of shows just how unsafe it was previously, and didn't need nearly CF speeds to get in when trying to place one.
Also, not part of your question here but related, but I think you overstate how good 3.02 din's actually was at long range (possibly because it slowing down makes it feel slower at reaching longer ranges than it really is), as well as the relevancy of such long range exchanges in general. Was curious myself exactly how quick they were to reach different distances, and went ahead and compared that as well:
Old din's only begins to reach somewhere faster when nearing full FD length spacing between players (~100% edge to edge length of BF and ~90% DL), which for the most part only really comes up in a theoretical full stage camp scenario to determine which player has less to gain by staying that far away and is going to want to prevent them from having that much space to begin with, and will never have trouble doing so, even if they're the least mobile character in the game against the fastest as it's simply way too much of any stage. Ultimately, whether the onus is on Zelda or the opponent to do this it comes back to ranges where old din's was too slow to safely use against a good portion of the cast anyway without already having some control by having existing ones up. In cases where they benefit both by keeping her at a distance and have the mobility/stage size to help them do that, having an improved midrange game becomes even more important for preventing them from getting around her and reclaiming stage position or punishing her attempts to do so, and being more able to quickly and safely while within a relevant range of them plant it to say aid in covering the platform option or something helps there.
3) Talked about this in above reply to Karmaic Avidity ^^
4) Why have supersweetspots on them in the first place =P They could have been reworked to be not quite as extreme in KB/hitlag/etc, but at that point why even keep them when there was a supersweetspot, a sweetspot, a... tartspot(?), and a flub on a 4 frame duration static kick animation, and the supersweet and sweet then become more similar. Many felt it added needless variance with its half the size of rest hitbox that had as much to do with how noodly the target character is as it did user precision to make contact with it, and missing often just meant landing a hit that's also quite powerful itself. As just my own opinion, as someone who never touched Zelda in brawl there's some bias for melee kicks as I like having the full strength on spaced kicks for the many characters she doesn't have good setups into deep kicks, and also prefer the feel of the high but still fluid hitlag.
Thank you very much for the response. The gifs are fantastic for backing up your points. I don't expect you to have a chance to respond again, so this is more for other people to talk about if they wish.
1) I understand your point that even Link boomerang is clankable and clankable by shield. I also understand that the intention was to remove this because it was never intended in the first place. Nevertheless, Zelda's "projectile" holds the distinction of not moving, which makes it unable to pressure an opponent on its own. Coupled with the fact that Zelda is slow, I think it reasonable to give the Din an exception to this universal clank rule. Not make it go through shield, of course, but not have its hitbox removed either.
2) Zelda was rather close to CFal in the first place in your gif, so I would not have done a retreating Dins either. I would have DD'd myself and spaced normals. I think the scenario when a retreating dins was valuable was when Zelda was a little further away, and the opponent was not quite as fast as CFal.
With regards to Dins travel speed, I was not saying that it reaches locations sooner than old Dins, what I was saying was that it does so 'at a quicker velocity'. This is evident even in your gif. With old Dins, if an opponent stood on the other side of the stage, I could swap between placing the Dins just outside their shield to block movement and placing it on their body as harass at the last second, and the opponent would have to predict to react to it. The new Dins is moving far too slowly to mix it up.
3) I agree about Nayru's aerial being different from grounded being a good thing, and I think the autocancel is a nice way to give her an offensive option. I still don't see why the invuln had to be outright removed though. Could you not have turned it into light armor? Reduced the invuln frames to a very short period? There are niche uses for nair and aerial nayrus now, but I still feel they are too similar.
Also, what happens if you land an aerial nayru's before the invuln frames would appear on the grounded version? Do you get invuln frames?
4) I can certainly live with changing the hitboxes. I can certainly live with making the first frame of the sweetspot do more damage/knockback akin to the supersweetspot. The problem is I don't get the feedback (as either player, or an audience) that the badass supersweetspot was hit. The hitlag on that thing was sexy, and having everyone know you hit a miniscule 1-frame hitbox just felt great. I think it was just unnecessary to remove it. I'm not sure what you're talking about with "noodly" characters, though. You mean ones that dont accidentally also hit the sourspot?
the changes proposed by magus have remedied my problems with zelda. (most any other things i have a problem with are justified changes)
as a recap, correct me if i'm wrong or missed something, these are the things we can expect from the patched zelda.
- din detonate will make the moving din hitbox disapear first, keep moving a little then stop and detonate so the detonate hitbox will be easier to hit.
- din will get a deactivated indicator
- up-b cancel will be able to reverse direction from a grounded version
- up-b cancel will have -2 frames cool down
- side b din to neutral b nayru can be mashed out. with side b being a specific detonate trigger.
- din may possibly get a shorter minimum cast distance
Thank you magus for your great responses! It's wonderful to finally get some answers on development questions we've been having.
I'm noticing an overall design trend with Zelda to shy away from ensuring real functionality when it requires a better than average mechanic/aspect/property. Not so much on the big, noticeable things like Din's Fire and Teleport, but in small details on her normals. My post will be mainly me attempting to point this out instance by instance, after this introduction explaining my perspective and thesis.
Thesis: Zelda is on the butt end of a double standard compared to the rest of the cast; any strengths she has are always under constant scrutiny, while those of others (whether for historical precedent or for sheer popularity) are permitted. To use an extreme for illustration of this 'trope,' look at some of Nintendo's decisions in the official smash games: 1) melee dins fire dealing only moderate knockback and damage combined with a small hitbox, extremely delayed hitbox, extremely slow start up, extremely low maneuverability, long endlag, the aerial version sending her into a freefall, 2) melee and brawl teleport traveling a long distance quickly but also long start up, very long landing lag, poor maneuverability during freefall 3) grab being a little larger and longer lasting than normal but being much slower than normal.
Now, I believe these choices were made by Nintendo to avoid having something be annoying, and that
these arecompensations to counterbalance their utility and prevent Zelda players from spamming things like these to win. The double standard lies in that other characters in these games have tools that have much fewer counterbalances to them (the obvious example is shine). I suspect in Nintendo's mind, shine wasn't expected to be annoying because it's difficult to use and therefore wouldn't be used effectively often. It isn't obvious to Nintendo that shine is what most of us would consider 'broken' because they weren't thinking from a competitive standpoint but purely about fun, whereas the PMDT's Zelda double standard results from a different perspective
The PMDT has a goal in mind: to improve the competitive gameplay with a few thematic guidelines, so
they view everything through the lens of balance. It's clearer to the PMDT that shine is a broken move and that difficulty of execution does not necessarily counterbalance it (you can tell because they removed the invincibility on shine as a token gesture), but their reason for permitting shine isn't because it requires technical skill but because of historical precedent and not wanting to alienate spacey players. Yet they continue to scale back tools on other characters that provided them with similar ways to gain reliably an advantage, moves that were no where near as unchecked as shine was, such as nayru's love. This touches on a subject I'm not going to discuss in length but think is worth mentioning here, that previously, one of the core strategies for balancing PM revolved around spacies; every character had to be able to handle spacies and have extremely good special mechanics to compete with them. This resulted in the 'fat' that the PMDT is trimming now, and the current re-emerging dominance of the melee top tiers. But in any case, it's a double standard.
The last point I want to make in this introduction is about the trending reasoning behind changing things:
'cause melee' This is, to say, the redesigning of moves and properties to closer match what they were in melee simply to make them more like melee (rather than changing them because there's an issue and melee just happens to be a close working model that solves it). Originally, PM was marketed as being a mixture of melee and brawl, but over the years it seems increasingly to be the case that grand design changes on PM have been made more or less 'cause melee', and while some people might take issue with that choice, there isn't anything wrong with that from a design perspective. With anything creative, if you scale out far enough, every artistic decision becomes arbitrary, but when you narrow it down to how a specific move works, 'cause melee' is far too general of a design goal to be the sole reason behind a change, and since its presence as a reasoning always requires other supporting reasons, why use it at all? This is why I'm always suspicious when the only understandable reason behind changing something is 'cause melee'.
magus said:
The largest changes in 3.5 on it however aren't from the angle but rather a global mechanic change that fixes a bug that made any non-tumble hit not cancel stun on landing if they were hit more than once in the air
OK, I didn't realize this. Very interesting! I suppose then if the jab were reverted back to a 55 degree angle it still wouldn't have the same function as before? I see what you mean about needing to hit deeper with the jab to get a similar combo function, but the jab is just soo slow. It used to be that its great disjoint was the treat for having such a slow jab, but now if you have to be within a normal jab's range to get a decent chance of a followup it just isn't worth doing, especially since at closer ranges your opponent can just shield grab you.
magus said:
a jab 2 concept that I ended up scrapping (if you're curious, very long range jab that I couldn't get to look decent and had high potential for stupid/completely eliminating a previously large weakness rather than lessening it)
It sounds like you had a similar notion and might agree with me here about disjoint being the compensation for sluggishness. And regarding your idea about adding a 2nd jab link-in for her, Zelda's jab is already more like a jab 2 than it is a single jab so why is it so weird to have the jab launch the opponent slightly more like Donkey Kong's jab 2 so that she can combo more reliably off of it. On the flip side, you could try the opposite direction and try giving her a jab 1 and making the current jab the jab 2 link-in. She could flick her right hand in front of her for a frame 4 jab and if you press A again she does the second jab, allowing for a jab combo. In any case, when you explain it, i understand the reasons that lead to the current change and I wouldn't necessarily disagree, but current jab needs some love now. I don't think every move needs to be the best, but it should be worth doing sometimes, and currently it just isn't when ftilt is better and faster for combos, nayru's love is much faster for a gtfo move, uptilt is faster and perhaps even more disjointed in a way and kills... you get my drift.
magus said:
Zelda having strictly one of the best shieldgrabs in the game just doesn't feel right to me, with her historically having abysmal grab speed but better non-grab options than most.
This is just a 'cause melee' excuse. She already diverges from past incarnations, why bother putting your foot down on preserving this particular property of hers when all it does is leave it in a strange half-state of normalized vs special grab, especially when you already took away what made it a special grab.
magus said:
I went with the frame 8 grab (same as DK) because her range is pretty high up there in the cast and her friction is high as well
I don't understand why she can't be allowed to have one of the better grabs. Double standard. Also, saying that her friction is too high to permit having a normal grab feels like a cheat. Her friction is a wide standing attribute that already has it's own pros and cons; it shouldn't count as a double whammy for counterbalancing actual moves. And honestly its range really isn't all that great. Its range is nothing like a tether grab, and it's not so big that it merits any special kind of compensation. Here's some data on 3.02 character's grab sizes: http://smashboards.com/threads/p-m-3-0-statistics-list-shff-speed-weight-falling-speed-etc.335019/ note that these do not include tether grabs, so adding those 8 now puts her at having the 21st (out of 41) biggest range on standing grab; again, not what I would consider large enough to merit special compensation in making it slower.
magus said:
She has among the very best long range OoS options on both sides for things normally safe due to spacing with her kicks, a faster than normal shieldgrabs overhead and very short horizontal range u-smash OoS, and sometimes helpful up-b OoS that hits on 7 and covers the space inside of kick range
There are only 2 other characters who have slow non-tether grabs: DK and bowser.
With bowser, I imagine having a really good OoS option with Up B is the reason behind keeping his grab
slow, but Zelda's OoS options have nothing anywhere near in comparison to the utility that bowser's Up B does (invicibility on frame 1 with hitboxes on frame 5 that cover his whole body). Yes in ways, her kicks are good OoS, but your opponent has to be exceptionally far from her to land the sweetspots, fair being frame 9 is somewhat fast but really not fast for an OoS option, and Bair requires the opponent to have crossed up your shield very far, and the sweetspot mechanic makes it extremely dangerous to risk whiffing since the higher priority flub hitboxes at most percents give your opponent frame advantage.
With DK, I imagine having the incredible conversions he gets off of grabs is the reason behind keeping it
slow, but again Zelda's conversions off of grabs are fine, but nothing like the particularly rewarding conversions that DK has. Unlike DK, when the opponent is grabbed they always have a way to escape a followup with good DI and Zelda just happens to have very slow throws so it's very easy to avoid her throw mixups, compared to DK's throws, which are fast. So, if there's any thought that having a normal grab will turn her into sheik, I really don't think that will be the case because of her merely average conversions off of grabs.
magus said:
The reach wasn't reduced by any meaningful amount though
(there are some tiny differences after some collisions were shifted around a bit). The collision edits were mainly reducing the inside ones on the body while leaving the exceptionally large ones at the hands
The change to the inside ones is a good change since it allowed her to sometimes grab people slightly behind her, which didn't make sense. However, the large one at her hands (grabbox ID 0) were in fact reduced from 4.89 to 4.69. So if it being too large before was an issue, you've already reduced it so why does it still need extra compensation in being slow?
There's actually a double standard within the vacuum of this move's redesign: the special properties of
the grab (having an lingering frame) were removed to be normalized yet the speed is still 1 frame slower than normal.
Finally there's still another bad property to zelda's grabs that already compensates for it having some
slightly better than average mechanics: she's so tall that people can crouch under it. Also, you removed the positioning of the hitboxes on pivot grabs so people can crouch under her grab even more. Once again, I really don't think her grab is so good that it merits special compensation.
My bad, I meant to say exiting not existing. I thought I remembered a grounded teleport taking until frame 33 to disappear before but now she disappears on frame 24. Double checking the old code, i realized that I had forgotten to account for the frame speed modifier so she really did used to disappear on frame 24. Although, looking at the new code, the frame speed modifier isn't there, but since she still disappears on frame 24 I assume you must have hidden the frame speed modifier somewhere?
This also brings up the question of the speed of the ribbon hit (the hitbox on the starting teleport
animation). Before, the the existing frame speed modifier actually made it a frame 5 OoS option, unless my calculations/understanding was wrong. But I don't see anything about that in the changelog, so I'm just wondering if you can confirm this.
otheusrex said:
B) I really wish that exiting from the air was the same speed as from the ground
B) To clarify, which parts are you referring to? In the current release it's:
Ground Startup: Hit: 6-7; Total: 24
Air Startup: Hit: 7-9; Total: 32
Travel: Total: 19
Ground Ending (from ground/air): Hit: 1-2; Total: 21
Ground Cancel (from ground): Total: 2 invisible + 12 (supposed to be 2+10)
Ground Cancel (from air): Total: 2 invisible + 21 (undecided about making this 2+19 or not since it makes sense to have the part with animation match the normal 21 frame ending animation if anything)
Landing Lag (from air reappearance then landing): 30
I mean that I wish the ground and air startup were the same speed, meaning that she disappeared on frame 24 so that the timing was the same between them. I think that 24 frames is still more than a punishable amount, and since you already added even more landing lag to teleport, it would be nice to just have the timing for it be consistent all the way around.
magus said:
Going with hitbox sizes instead would make the larger ones unavoidably easier and to a larger degree as well since they all share the same hitbox data and are just scaled automatically by model size
Which hitboxes are you referring to exactly? as there are 3 different types of explosions on the dins fire: a) the full explosion if you let the timer on the mine run its full course, b) the manual detonation explosion if you press side b after you've set a mine but before it's timer has run out, or c) the wisp explosion if you press side b while the dins fire is returning to you.
I believe Drseusredfish was referring to the manual detonation explosions size, but your response to his
question--
magus said:
I came up with the idea of instead having it activate this sequence: flame goes out for a couple frames just before detonating -> stop + explosion -> re-ingnite, instead of the current stop + explosion -> re-ingnite. It'd be moving a little before the explosion without the return hitbox being there to hit them, so on uncharged ones where the space between the return hitbox and explosion is smallest it'd be compensated by the faster travel speed, while large ones would stay mostly the same. Going with hitbox sizes instead would make the larger ones unavoidably easier and to a larger degree as well since they all share the same hitbox data and are just scaled automatically by model size.
your response to his question suggests to me you were only addressing the concerns about the wisp explosion--and that seems like a decent idea to solve that issue-- but I want to underscore drseusredfish's concerns about the manual explosion. It really is too small a difference of area for you to hit an opponent that was already going to hit the mine. I would prefer it if the manual detonation size were just the same size as the full explosion but with the same lessened knockback of the current manual detonation.
I suppose your proposal above could also be adapted to solve this issue too by having the mine hitbox
disappear for a moment before your detonation, but if you're going to do that I'd also suggest upping the base knockback. The reward for your opponent hitting the mine vs the explosion is the same (and unless I'm mistaken EXACTLY the same), so it doesn't make sense from a design standpoint to have a more difficult option have the same reward as the easier one. At the least, it should have a different function. Also, I'm suggesting base knockback instead of knockback growth because I assume you wouldn't want the manual detonation to become a kill move.
Finally, in regards to
magus said:
they all share the same hitbox data and are just scaled automatically by model size
I know that the size of the full explosions are controlled by two different floating points: 16ab4 (intial hitbox size multiplier), and 16ab8 (terminal hitbox size multiplier). Because of this, you could make the size of the close and small explosions bigger while preserving the size of the biggest explosions by scaling up one floating point and scaling down the other. I imagine when engineering the new dins you had to invent a new hitbox formula for it, but I don't imagine that it's out of the range of possibility to code something close to this arrangement if you wanted to make the small explosions bigger without affecting the larger ones.
In summary, Zelda is subject to an unfair double standard in that designers typically counterbalance strengths on her moves more than they do on other characters. This is also what I like to call Constant Counterbalance Syndrome, i.e. the idea that every strength needs an equal or comparable weakness in order to be balanced, and is probably the reason why she has always started out seeming like a good/decent/average character in smash games at first, only to slide further and further down the tier list as people discover the other character's exploits. Specific examples of her double standard include nayru's phobia compared to the other poster child defensive moves, shine vs nayru and counter vs nayru, where the former comparison sees people complaining that a defensive move is used as a bread n butter combo tool and therefore can't be right somehow, and the latter comparison sees people complaining that by a simple button press and correct timing you negate an opponent hitting you and turn it into a full reversal, sometimes even leading to kills. A specific example of the double standard applied to 3.5 zelda changes is the treatment of her grab, where fire emblem characters have more disjointed grabs that are normal speed but Zelda's slightly larger than average(21st largest out of 41) grab needs to be frame 8 to compensate.
Truly, I suspect that the biggest reason why people overcompensate when balancing Zelda is that they're
just used to her being bad. Zelda's weaknesses were so large before that they required a great amount of improvement to be overcome. When these changes are examined, they seem suspect, due to the drastic improvement not seen in other characters. Pikachu, Samus, Luigi, etc., only required a few small buffs to compete more reliably, but Zelda, Bowser, Mewtwo, etc., were buffed enormously. I'd also like to point out another double standard comparing Mewtwo and Zelda in that both were lowest tier characters who needed great improvements, but mewtwo was permitted to have extremely broken mechanics to ensure a high-tier level of viability, whereas Zelda's was simply buffed to be passable. Yes, I realize that mewtwo was nerfed and isn't as good as he used to, but if you compare many of his properties to those of other characters, he still has many attributes that are just superior to other characters, and people are more accepting of mewtwo being superior to other characters than Zelda.
'Cause Melee' is a mentality that's behind a lot of changes recently. It's fine for an overarching design
goal, but problematic when it becomes the driving factor in specific move changes to exclusion of other criteria. It makes sense for design changes to be made to emulate how something was in melee with characters who functioned well in melee, but why would this strategy work for Zelda when she was the 3rd worst character in it because her moves don't work in melee. This is specifically why I disagree with the jab change. In general, her functionality is being sacrificed for reasons that don't have to do with balance, and 'cause melee' just isn't a reason to change something unless all other factors are equal; it just isn't good design.
Once again, magus, I want to thank you for being here to answer some of our questions. I know that some of the things I say might seem harsh, but I just want you to know that I really appreciate the effort you put into making a new Zelda build and all the hard work you did in coding and designing everything. I was expecting a Zelda from the PMDT that would be simply watered down and more fun for people to fight against because she was easier to beat, but it comes across to me that you tried to make a new Zelda build that would address the PMDT's concerns but would still be fun to play as. I hope I didn't seem to be personally criticizing or attacking you at any point here, and I once again, await your response eagerly.
2) ...With regards to Dins travel speed, I was not saying that it reaches locations sooner than old Dins, what I was saying was that it does so 'at a quicker velocity'. This is evident even in your gif. With old Dins, if an opponent stood on the other side of the stage, I could swap between placing the Dins just outside their shield to block movement and placing it on their body as harass at the last second, and the opponent would have to predict to react to it. The new Dins is moving far too slowly to mix it up.
Ah ok. From that far away though along with the din's always moving forward to some degree regardless of angle and the delay before the hitbox comes out after stopping, they really shouldn't be getting caught by that I don't think since they're completely safe to just roll forward or something as it would begin hitting if placed where they are (and stopping it early enough to hit where a roll would go would be far enough apart in time to see it and not do the roll), then ignore or clank it if it's close enough to be a threat. Takes a good 1.5 or more seconds from start to finish at that distance with either din's so they'd have time to reset the situation as you throw the next if nothing else, though tbh I don't remember the last time I actually played out that scenario with neither player thinking they should be staying closer in. With the new one though you might get to detontate it on them when they try to hit it if placed above shield level or at least have a large one out there to try to play off of.
4) I can certainly live with changing the hitboxes. I can certainly live with making the first frame of the sweetspot do more damage/knockback akin to the supersweetspot. The problem is I don't get the feedback (as either player, or an audience) that the badass supersweetspot was hit. The hitlag on that thing was sexy, and having everyone know you hit a miniscule 1-frame hitbox just felt great. I think it was just unnecessary to remove it. I'm not sure what you're talking about with "noodly" characters, though. You mean ones that dont accidentally also hit the sourspot?
I can see the appeal there as a finisher, but when connected outside of a kill I personally felt it mostly just made them feel rather clunky. Some other attacks like rest, falcon punch, luigi up-b, full charged eruption, etc have gigantic hitlag for flavor, but they're pretty much only going to connect when they're going to kill something or close to it. From a functional standpoint the hitlag also means they're never going to DI it poorly like can happen with the normal sweetspots on occasion when they weren't expecting to be hit, and the >1x hitlag multiplier makes the shield advantage worse since shields always take 1x hitlag regardless of multiplier while Zelda sits in the 1.4x hitlag as if it hit. Super hits were -2 on block vs the current sweetspots' +1 despite doing less damage.
By noodly characters I mean small hitboxes like the super toe needed have unreliable z axis coverage for long slim hurtboxes on arms/legs moving around yet they can get in the way to be hit by the flub when aiming for the larger, more stationary, and centered body hurtboxes (unlike something like rest where it's just that 1 tiny hitbox to be concerned with). It was significantly easier to land the super on fat characters without long slender hurtboxes complicating the spacing, while larger hitboxes don't have issues connecting consistently on any part of the body with the filled out z coverage.
----------------------------------------
OK, I didn't realize this. Very interesting! I suppose then if the jab were reverted back to a 55 degree angle it still wouldn't have the same function as before? I see what you mean about needing to hit deeper with the jab to get a similar combo function, but the jab is just soo slow. It used to be that its great disjoint was the treat for having such a slow jab, but now if you have to be within a normal jab's range to get a decent chance of a followup it just isn't worth doing, especially since at closer ranges your opponent can just shield grab you.
It sounds like you had a similar notion and might agree with me here about disjoint being the compensation for sluggishness. And regarding your idea about adding a 2nd jab link-in for her, Zelda's jab is already more like a jab 2 than it is a single jab so why is it so weird to have the jab launch the opponent slightly more like Donkey Kong's jab 2 so that she can combo more reliably off of it. On the flip side, you could try the opposite direction and try giving her a jab 1 and making the current jab the jab 2 link-in. She could flick her right hand in front of her for a frame 4 jab and if you press A again she does the second jab, allowing for a jab combo. In any case, when you explain it, i understand the reasons that lead to the current change and I wouldn't necessarily disagree, but current jab needs some love now. I don't think every move needs to be the best, but it should be worth doing sometimes, and currently it just isn't when ftilt is better and faster for combos, nayru's love is much faster for a gtfo move, uptilt is faster and perhaps even more disjointed in a way and kills... you get my drift.
On non-floaties it wouldn't work the same as before with a 55 angle yeah. Difference between 361/55 mostly just affects the default KB not funneling right into standing f-tilt's range, and the dash grab/attack timings out of it not being as lenient. It's still her shortest total duration (23) and lowest endlag move, however. Stating the obvious here but the range is good, low commitment, safe on whiff, and nice on block for a ground move (-5) even if you can get grabbed if used in their face (though startup of f-tilt animation can dodge-counter a shield grab if near the outside of their reach). If you want an attack to put out in case they get caught by it while remaining relatively safe if they don't it's a good option. It's not a great move no, but it's not overshadowed by other options in the areas it excels in either though like you're saying, which was never really its ability to combo at low damage if they don't hold down or away.
Her jab isn't more like a jab 2 or anything else really. Its range + very low cooldown at the cost of extra startup on something between jab/fast tilt total length is pretty unique in general. I don't know what I'd best try to compare it to, but it doesn't really have much of anything in common with DK's jab 2 other than being in the jab moveslot. I'd probably look at increasing the KBG first if wanting to improve it in some way so DI into the floor knocks down earlier than 60 something before making it combo better if they don't, or maybe a little of both since more KB would make that a bit worse.
This is just a 'cause melee' excuse. She already diverges from past incarnations, why bother putting your foot down on preserving this particular property of hers when all it does is leave it in a strange half-state of normalized vs special grab, especially when you already took away what made it a special grab.
I don't understand why she can't be allowed to have one of the better grabs. Double standard. Also, saying that her friction is too high to permit having a normal grab feels like a cheat. Her friction is a wide standing attribute that already has it's own pros and cons; it shouldn't count as a double whammy for counterbalancing actual moves. And honestly its range really isn't all that great. Its range is nothing like a tether grab, and it's not so big that it merits any special kind of compensation. Here's some data on 3.02 character's grab sizes: http://smashboards.com/threads/p-m-3-0-statistics-list-shff-speed-weight-falling-speed-etc.335019/ note that these do not include tether grabs, so adding those 8 now puts her at having the 21st (out of 41) biggest range on standing grab; again, not what I would consider large enough to merit special compensation in making it slower.
There are only 2 other characters who have slow non-tether grabs: DK and bowser.
With bowser, I imagine having a really good OoS option with Up B is the reason behind keeping his grab
slow, but Zelda's OoS options have nothing anywhere near in comparison to the utility that bowser's Up B does (invicibility on frame 1 with hitboxes on frame 5 that cover his whole body). Yes in ways, her kicks are good OoS, but your opponent has to be exceptionally far from her to land the sweetspots, fair being frame 9 is somewhat fast but really not fast for an OoS option, and Bair requires the opponent to have crossed up your shield very far, and the sweetspot mechanic makes it extremely dangerous to risk whiffing since the higher priority flub hitboxes at most percents give your opponent frame advantage.
With DK, I imagine having the incredible conversions he gets off of grabs is the reason behind keeping it
slow, but again Zelda's conversions off of grabs are fine, but nothing like the particularly rewarding conversions that DK has. Unlike DK, when the opponent is grabbed they always have a way to escape a followup with good DI and Zelda just happens to have very slow throws so it's very easy to avoid her throw mixups, compared to DK's throws, which are fast. So, if there's any thought that having a normal grab will turn her into sheik, I really don't think that will be the case because of her merely average conversions off of grabs.
Then why reduce it at all?
The change to the inside ones is a good change since it allowed her to sometimes grab people slightly behind her, which didn't make sense. However, the large one at her hands (grabbox ID 0) were in fact reduced from 4.89 to 4.69. So if it being too large before was an issue, you've already reduced it so why does it still need extra compensation in being slow?
There's actually a double standard within the vacuum of this move's redesign: the special properties of
the grab (having an lingering frame) were removed to be normalized yet the speed is still 1 frame slower than normal.
Finally there's still another bad property to zelda's grabs that already compensates for it having some
slightly better than average mechanics: she's so tall that people can crouch under it. Also, you removed the positioning of the hitboxes on pivot grabs so people can crouch under her grab even more. Once again, I really don't think her grab is so good that it merits special compensation.
...
A specific example of the double standard applied to 3.5 zelda changes is the treatment of her grab, where fire emblem characters have more disjointed grabs that are normal speed but Zelda's slightly larger than average(21st largest out of 41) grab needs to be frame 8 to compensate.
As far as grab stuff goes I'm just going to clear up some things first, since it seems like much of your stance is based off of misinformation, because from my perspective you appear kinda ridiculous to me here tbh to be taking such strong issue with the current state of her grab and describing it as some sort of anti-Zelda bias :x
1) The 3rd active frame never existed outside of PM. It was a throwaway 'buff' to make her grab technically not be as terrible when it was still starting on 10-12 or something. I probably stuck that on there years ago myself before or after I made it faster than the original 12. On a character with a bad JC grab slide vs dash grab pretty much the only thing that ended up doing for her is making sloppy timing on chaingrabs not miss by grabbing too early. Any other time that 3rd standing frame would come up in play would be extremely rare (delayed shieldgrab + they spotdodged or something at just the right time). The 1 frame faster startup does far more, and even that isn't too significant of a change overall.
2) That grab range list you're looking at is complete garbage. Not only does it attempt to rank the range from TopN (base bone of model usually at bottom center) to the outermost edge of the grab collisions, a pretty useless metric in many cases, there are no actual numbers shown and the rankings themselves aren't even correct! The method listed is what's supposed to be done there (offset + radius), but how they came to that ranking order I haven't a clue, because it's never been accurate in any version of the game and I think I mentioned it somewhere in that thread. Here is a correct listing of that TopN range measurement:
3) Dumping the tether characters in front of her to somehow suggest her range isn't as good as I was saying and is instead only average is completely silly and I really hope you are not serious there. They all have 50%+ more startup at short range, often around double that startup to actually reach full length, and double the endlag. They're different beasts entirely and not even remotely comparable when talking about OoS capability.
4) As I mentioned in 2, measuring the distance from TopN to the edge of the grab says very little about the effective reach of the grab. It doesn't take into account their size and hittable area, as what matters a lot more is the distance between the spacing to normally hit the grabber and where their potential grab can extend to. For example, Bowser is comfortably sitting at #3 in that measurement, but his reach barely extends beyond his face. Even if it grabbed on 7 it wouldn't be anywhere near as good as the list suggests it is.
...So here's this effective grab reach comparison I made instead that shows the upper half of the non-tether list in a more meaningful way:
Zelda's slim frame puts her all the way up at 4th longest reach (going by shield animations isn't much different if curious, and if anything she'd likely move above Roy since Marth/Roy stick their leg out more). Probably close to 1st with shield grab range when you consider how much the others slide (yes it matters when talking about OoS options). Nowhere near average.
5) Her kicks being great OoS at long range was entirely why I mention them! How many other characters have punishes at that range on block at that speed with anywhere near their power? Pretty much 0. Sure, they are slower than normal shield grabs, but that's not why you'd use them. They can hit things well beyond even Marth's grab range on both sides and even f-air is faster than any character's wavedash OoS speed before even considering the attack after it. Shield DI also helps here some in 3.5 to get the spacing you need for the sweetspot, and with a fully retreating jump and c-sticked you can connect them from surprisingly close range as well.
A lot of shield interaction involves spacing outside of their OoS options rather than simply outspeeding their grab or whatever, particularly on ground moves and with slower and/or ranged characters. Bowser has inv startup on up-b OoS yes, but the strongest hit also has negative reach so you can space moves on his shield to pretty good effect instead. Zelda on the other hand excels against these attacks that aren't safe up close, but take significantly longer with most of the cast to reach OoS with a punish to make them safe on block against them.
Wasn't really my idea to change it from 9 in the first place actually as I felt it was fast enough to be usable while keeping the choice between her other OoS options more apparent, and leaving a little soft spot in her OoS capability at an unusual range. She's great at long range OoS from either side (moreso on back with b-air's startup/able to hit lower), about as good as non-inv startup close range options get with u-smash, and had a good range but noticeably slower grab than average that a low reward up-b can cover between retreating toe/u-smash range if that bit of speed difference is needed or on the back side. As her grab approaches 7 that dynamic and unique bit of weak OoS range goes away.
6) Turn grab is still able to grab all of the lowest crouches in the game just fine. I made sure of it when I adjusted it and the other tall standing grab characters' dash/turngrabs in 3.5 to have an option available to grab them with.
Can you see where I'm coming from here? It's not overcompensation and bias against Zelda, I see it as sensible design in my eyes and I can't really think of a character that has such solid coverage to various angles and ranges of shield pressure as what you're asking for. Generally if they have a stronger short/inside range option than her they lack in the longer ranged/sub-WDOoS speed options. If you still feel it's unfair for her standing grab to not come out on frame 7 we'll have to just disagree, as she has more going for her from shield than most characters imo regardless of whether it comes out on 7-10.
My bad, I meant to say exiting not existing. I thought I remembered a grounded teleport taking until frame 33 to disappear before but now she disappears on frame 24. Double checking the old code, i realized that I had forgotten to account for the frame speed modifier so she really did used to disappear on frame 24. Although, looking at the new code, the frame speed modifier isn't there, but since she still disappears on frame 24 I assume you must have hidden the frame speed modifier somewhere?
This also brings up the question of the speed of the ribbon hit (the hitbox on the starting teleport
animation). Before, the the existing frame speed modifier actually made it a frame 5 OoS option, unless my calculations/understanding was wrong. But I don't see anything about that in the changelog, so I'm just wondering if you can confirm this.
I mean that I wish the ground and air startup were the same speed, meaning that she disappeared on frame 24 so that the timing was the same between them. I think that 24 frames is still more than a punishable amount, and since you already added even more landing lag to teleport, it would be nice to just have the timing for it be consistent all the way around.
The FSM is just moved from the .pac script into the codeset. FSMs in subactions are reset on subaction change, like when going from ground <-> air, so being in the codeset's animation engine to do the FSM makes it always be active while on the ground in case you land during startup. I did this with some nayru's/din's FSMs as well, since specials shouldn't have subaction FSMs while they're able to switch between air/ground animations. Hitbox speed is unchanged (6), other than if you land right after starting it it will come out faster than previously because the FSM will be properly applied on landing.
Even though making air/ground speeds match would be nice from a consistency standpoint, I really don't think her recovery needs any more improvement than it's already received with being able to shorten it, able to land/edgecancel on platforms while moving horizontally or upwards for the much safer ground reappearance, and her improved airdodge. Reducing startup mostly just hurts the characters that can't fly offstage quickly to hit her during startup. If they start to go off to try to intercept you could safely get moving before they can reach you from much closer to the stage, and with being able to safely start from closer means when they instead need to stay at the edge to be able to attempt to punish it you have much greater options, especially if there's a platform, with one of them most likely being unpunishable from their position when you go to aim it or at least require predicting exactly what you're going to do (ground reappear somewhere between full onstage and at edge, go for edgegrab, ground reappear/edge cancel on platform with cancel, etc).
If you are forced into reappearing in the air to eat the reappear + 30 landing lag on recovery you most likely messed up, either with bad DI or were intercepted far offstage. She doesn't need to even use up-b to reach the stage in most cases with her air speed, let alone need to up-b from below stage height where you can't aim it to end on the ground for low endlag and they can hold the edge and have time to punish the landing lag unlike the ground reappearance that would require a read of some sort between it and going for the edge (though with a platform there you can still get the ground reappearance anyway by using the cancel snap onto it). 3/4 startup and the very much avoidable higher air to ground lag aren't anywhere near comparable really.
Which hitboxes are you referring to exactly? as there are 3 different types of explosions on the dins fire: a) the full explosion if you let the timer on the mine run its full course, b) the manual detonation explosion if you press side b after you've set a mine but before it's timer has run out, or c) the wisp explosion if you press side b while the dins fire is returning to you.
I believe Drseusredfish was referring to the manual detonation explosions size, but your response to his
question--
your response to his question suggests to me you were only addressing the concerns about the wisp explosion--and that seems like a decent idea to solve that issue-- but I want to underscore drseusredfish's concerns about the manual explosion. It really is too small a difference of area for you to hit an opponent that was already going to hit the mine. I would prefer it if the manual detonation size were just the same size as the full explosion but with the same lessened knockback of the current manual detonation.
I suppose your proposal above could also be adapted to solve this issue too by having the mine hitbox
disappear for a moment before your detonation, but if you're going to do that I'd also suggest upping the base knockback. The reward for your opponent hitting the mine vs the explosion is the same (and unless I'm mistaken EXACTLY the same), so it doesn't make sense from a design standpoint to have a more difficult option have the same reward as the easier one. At the least, it should have a different function. Also, I'm suggesting base knockback instead of knockback growth because I assume you wouldn't want the manual detonation to become a kill move.
Finally, in regards to
I know that the size of the full explosions are controlled by two different floating points: 16ab4 (intial hitbox size multiplier), and 16ab8 (terminal hitbox size multiplier). Because of this, you could make the size of the close and small explosions bigger while preserving the size of the biggest explosions by scaling up one floating point and scaling down the other. I imagine when engineering the new dins you had to invent a new hitbox formula for it, but I don't imagine that it's out of the range of possibility to code something close to this arrangement if you wanted to make the small explosions bigger without affecting the larger ones.
There are only 2 explosions not 3: Timed and manual activation. As far as adding a 3rd one specific to activating it before it's moving goes, giving immediate on call access to the full sized explosion that's nearly as large as Bowser at minimum even if it's weak sounds like a pretty bad idea to me. Maybe something a little larger/stronger than the re-activation explosion (would probably just up the damage a bit which would also increase KB), but the opponent is supposed to be able to get close enough to these things to attempt to hit them without getting blown up too easily.
Mine hitbox does not have the same KB as the manual explosion. The explosions are more BKB heavy compared to the stationary/return hitboxes.
Those 2 din's special floats you mentioned currently do nothing actually, as they're the child article hitbox scaling, and to make the din's able to go into hitlag I forgo calling the child article entirely for the explosions and create them and the GFX/SFX within the parent article, as well as calculating the variable damages for them and the non-explosion hitboxes based on charge time. Even if that weren't the case though, those 2 floats should never be unequal to each other, as it makes the hitbox scale unevenly with the GFX size. I don't know why they had them not match in brawl originally (possibly just copied from parent scaling by mistake as they were 1.0/1.9 on both), but it's what made the GFX not match the hitbox and be too large or too small for where it hits depending on charge level, and those 2 floats have matched each other since before PM 1.0 to fix that issue.
There are 4 hitboxes, and all are scaled based on model size, which is controlled by the 16A68/16A6C floats (parent article scaling based on charge percentage). If return hitbox size is 2, manual 8, base size 1x, and max size 4x, that's a difference of 6 at min and 24 max. Any amount I adjust one of those hitboxes is only going to more significantly impact the larger charges was what I was saying. The explosion/non-explosion hitboxes and GFX are all linked by the parent article's model size.
Probably last reply of significant length from me for a while.
----------------------------------------
Ah ok. From that far away though along with the din's always moving forward to some degree regardless of angle and the delay before the hitbox comes out after stopping, they really shouldn't be getting caught by that I don't think since they're completely safe to just roll forward or something as it would begin hitting if placed where they are (and stopping it early enough to hit where a roll would go would be far enough apart in time to see it and not do the roll), then ignore or clank it if it's close enough to be a threat. Takes a good 1.5 or more seconds from start to finish at that distance with either din's so they'd have time to reset the situation as you throw the next if nothing else, though tbh I don't remember the last time I actually played out that scenario with neither player thinking they should be staying closer in. With the new one though you might get to detontate it on them when they try to hit it if placed above shield level or at least have a large one out there to try to play off of.
You should get the inv, but I don't think it's possible to land by frame 4 unless on an upwards slope because gravity is briefly disabled on startup.
I can see the appeal there as a finisher, but when connected outside of a kill I personally felt it mostly just made them feel rather clunky. Some other attacks like rest, falcon punch, luigi up-b, full charged eruption, etc have gigantic hitlag for flavor, but they're pretty much only going to connect when they're going to kill something or close to it. From a functional standpoint the hitlag also means they're never going to DI it poorly like can happen with the normal sweetspots on occasion when they weren't expecting to be hit, and the >1x hitlag multiplier makes the shield advantage worse since shields always take 1x hitlag regardless of multiplier while Zelda sits in the 1.4x hitlag as if it hit. Super hits were -2 on block vs the current sweetspots' +1 despite doing less damage.
By noodly characters I mean small hitboxes like the super toe needed have unreliable z axis coverage for long slim hurtboxes on arms/legs moving around yet they can get in the way to be hit by the flub when aiming for the larger, more stationary, and centered body hurtboxes (unlike something like rest where it's just that 1 tiny hitbox to be concerned with). It was significantly easier to land the super on fat characters without long slender hurtboxes complicating the spacing, while larger hitboxes don't have issues connecting consistently on any part of the body with the filled out z coverage.
----------------------------------------
On non-floaties it wouldn't work the same as before with a 55 angle yeah. Difference between 361/55 mostly just affects the default KB not funneling right into standing f-tilt's range, and the dash grab/attack timings out of it not being as lenient. It's still her shortest total duration (23) and lowest endlag move, however. Stating the obvious here but the range is good, low commitment, safe on whiff, and nice on block for a ground move (-5) even if you can get grabbed if used in their face (though startup of f-tilt animation can dodge-counter a shield grab if near the outside of their reach). If you want an attack to put out in case they get caught by it while remaining relatively safe if they don't it's a good option. It's not a great move no, but it's not overshadowed by other options in the areas it excels in either though like you're saying, which was never really its ability to combo at low damage if they don't hold down or away.
Her jab isn't more like a jab 2 or anything else really. Its range + very low cooldown at the cost of extra startup on something between jab/fast tilt total length is pretty unique in general. I don't know what I'd best try to compare it to, but it doesn't really have much of anything in common with DK's jab 2 other than being in the jab moveslot. I'd probably look at increasing the KBG first if wanting to improve it in some way so DI into the floor knocks down earlier than 60 something before making it combo better if they don't, or maybe a little of both since more KB would make that a bit worse.
As far as grab stuff goes I'm just going to clear up some things first, since it seems like much of your stance is based off of misinformation, because from my perspective you appear kinda ridiculous to me here tbh to be taking such strong issue with the current state of her grab and describing it as some sort of anti-Zelda bias :x
1) The 3rd active frame never existed outside of PM. It was a throwaway 'buff' to make her grab technically not be as terrible when it was still starting on 10-12 or something. I probably stuck that on there years ago myself before or after I made it faster than the original 12. On a character with a bad JC grab slide vs dash grab pretty much the only thing that ended up doing for her is making sloppy timing on chaingrabs not miss by grabbing too early. Any other time that 3rd standing frame would come up in play would be extremely rare (delayed shieldgrab + they spotdodged or something at just the right time). The 1 frame faster startup does far more, and even that isn't too significant of a change overall.
2) That grab range list you're looking at is complete garbage. Not only does it attempt to rank the range from TopN (base bone of model usually at bottom center) to the outermost edge of the grab collisions, a pretty useless metric in many cases, there are no actual numbers shown and the rankings themselves aren't even correct! The method listed is what's supposed to be done there (offset + radius), but how they came to that ranking order I haven't a clue, because it's never been accurate in any version of the game and I think I mentioned it somewhere in that thread. Here is a correct listing of that TopN range measurement:
3) Dumping the tether characters in front of her to somehow suggest her range isn't as good as I was saying and is instead only average is completely silly and I really hope you are not serious there. They all have 50%+ more startup at short range, often around double that startup to actually reach full length, and double the endlag. They're different beasts entirely and not even remotely comparable when talking about OoS capability.
4) As I mentioned in 2, measuring the distance from TopN to the edge of the grab says very little about the effective reach of the grab. It doesn't take into account their size and hittable area, as what matters a lot more is the distance between the spacing to normally hit the grabber and where their potential grab can extend to. For example, Bowser is comfortably sitting at #3 in that measurement, but his reach barely extends beyond his face. Even if it grabbed on 7 it wouldn't be anywhere near as good as the list suggests it is.
...So here's this effective grab reach comparison I made instead that shows the upper half of the non-tether list in a more meaningful way:
Zelda's slim frame puts her all the way up at 4th longest reach (going by shield animations isn't much different if curious, and if anything she'd likely move above Roy since Marth/Roy stick their leg out more). Probably close to 1st with shield grab range when you consider how much the others slide (yes it matters when talking about OoS options). Nowhere near average.
5) Her kicks being great OoS at long range was entirely why I mention them! How many other characters have punishes at that range on block at that speed with anywhere near their power? Pretty much 0. Sure, they are slower than normal shield grabs, but that's not why you'd use them. They can hit things well beyond even Marth's grab range on both sides and even f-air is faster than any character's wavedash OoS speed before even considering the attack after it. Shield DI also helps here some in 3.5 to get the spacing you need for the sweetspot, and with a fully retreating jump and c-sticked you can connect them from surprisingly close range as well.
A lot of shield interaction involves spacing outside of their OoS options rather than simply outspeeding their grab or whatever, particularly on ground moves and with slower and/or ranged characters. Bowser has inv startup on up-b OoS yes, but the strongest hit also has negative reach so you can space moves on his shield to pretty good effect instead. Zelda on the other hand excels against these attacks that aren't safe up close, but take significantly longer with most of the cast to reach OoS with a punish to make them safe on block against them.
Wasn't really my idea to change it from 9 in the first place actually as I felt it was fast enough to be usable while keeping the choice between her other OoS options more apparent, and leaving a little soft spot in her OoS capability at an unusual range. She's great at long range OoS from either side (moreso on back with b-air's startup/able to hit lower), about as good as non-inv startup close range options get with u-smash, and had a good range but noticeably slower grab than average that a low reward up-b can cover between retreating toe/u-smash range if that bit of speed difference is needed or on the back side. As her grab approaches 7 that dynamic and unique bit of weak OoS range goes away.
6) Turn grab is still able to grab all of the lowest crouches in the game just fine. I made sure of it when I adjusted it and the other tall standing grab characters' dash/turngrabs in 3.5 to have an option available to grab them with.
Can you see where I'm coming from here? It's not overcompensation and bias against Zelda, I see it as sensible design in my eyes and I can't really think of a character that has such solid coverage to various angles and ranges of shield pressure as what you're asking for. Generally if they have a stronger short/inside range option than her they lack in the longer ranged/sub-WDOoS speed options. If you still feel it's unfair for her standing grab to not come out on frame 7 we'll have to just disagree, as she has more going for her from shield than most characters imo regardless of whether it comes out on 7-10.
The FSM is just moved from the .pac script into the codeset. FSMs in subactions are reset on subaction change, like when going from ground <-> air, so being in the codeset's animation engine to do the FSM makes it always be active while on the ground in case you land during startup. I did this with some nayru's/din's FSMs as well, since specials shouldn't have subaction FSMs while they're able to switch between air/ground animations. Hitbox speed is unchanged (6), other than if you land right after starting it it will come out faster than previously because the FSM will be properly applied on landing.
Even though making air/ground speeds match would be nice from a consistency standpoint, I really don't think her recovery needs any more improvement than it's already received with being able to shorten it, able to land/edgecancel on platforms while moving horizontally or upwards for the much safer ground reappearance, and her improved airdodge. Reducing startup mostly just hurts the characters that can't fly offstage quickly to hit her during startup. If they start to go off to try to intercept you could safely get moving before they can reach you from much closer to the stage, and with being able to safely start from closer means when they instead need to stay at the edge to be able to attempt to punish it you have much greater options, especially if there's a platform, with one of them most likely being unpunishable from their position when you go to aim it or at least require predicting exactly what you're going to do (ground reappear somewhere between full onstage and at edge, go for edgegrab, ground reappear/edge cancel on platform with cancel, etc).
If you are forced into reappearing in the air to eat the reappear + 30 landing lag on recovery you most likely messed up, either with bad DI or were intercepted far offstage. She doesn't need to even use up-b to reach the stage in most cases with her air speed, let alone need to up-b from below stage height where you can't aim it to end on the ground for low endlag and they can hold the edge and have time to punish the landing lag unlike the ground reappearance that would require a read of some sort between it and going for the edge (though with a platform there you can still get the ground reappearance anyway by using the cancel snap onto it). 3/4 startup and the very much avoidable higher air to ground lag aren't anywhere near comparable really.
There are only 2 explosions not 3: Timed and manual activation. As far as adding a 3rd one specific to activating it before it's moving goes, giving immediate on call access to the full sized explosion that's nearly as large as Bowser at minimum even if it's weak sounds like a pretty bad idea to me. Maybe something a little larger/stronger than the re-activation explosion (would probably just up the damage a bit which would also increase KB), but the opponent is supposed to be able to get close enough to these things to attempt to hit them without getting blown up too easily.
Mine hitbox does not have the same KB as the manual explosion. The explosions are more BKB heavy compared to the stationary/return hitboxes.
Those 2 din's special floats you mentioned currently do nothing actually, as they're the child article hitbox scaling, and to make the din's able to go into hitlag I forgo calling the child article entirely for the explosions and create them and the GFX/SFX within the parent article, as well as calculating the variable damages for them and the non-explosion hitboxes based on charge time. Even if that weren't the case though, those 2 floats should never be unequal to each other, as it makes the hitbox scale unevenly with the GFX size. I don't know why they had them not match in brawl originally (possibly just copied from parent scaling by mistake as they were 1.0/1.9 on both), but it's what made the GFX not match the hitbox and be too large or too small for where it hits depending on charge level, and those 2 floats have matched each other since before PM 1.0 to fix that issue.
There are 4 hitboxes, and all are scaled based on model size, which is controlled by the 16A68/16A6C floats (parent article scaling based on charge percentage). If return hitbox size is 2, manual 8, base size 1x, and max size 4x, that's a difference of 6 at min and 24 max. Any amount I adjust one of those hitboxes is only going to more significantly impact the larger charges was what I was saying. The explosion/non-explosion hitboxes and GFX are all linked by the parent article's model size.
Again with the beautiful pictures. You're awesome.
I'm not sure what you're talking about in your response to Din's movement speed. Since I can't use what you said as a conversation point, I'll try to elaborate on what I said before. The movement speed of Din's as it neared the opponent allowed it to very suddenly change the area it threatened from in front/above the opponent to inside/above/behind the opponent. This often forced my opponent to sit in shield briefly, which gives me time to start placing a 2nd Din or begin approaching. At the very least, it ensured that the Din was not immediately clanked. I didn't mind if they rolled past it, as I still controlled the space for future exchanges.
Ok, I understand that the removal of the supersweetspot is mostly a good thing for Zelda. I still felt that the extra-powerful hit deserved feedback of some sort. If the hitlag is detrimental to Zelda (and forces the move to be un-DI-able for opponents), perhaps a visual/audio cue instead?
I think what Otherusrex is saying is that Zelda gets the **** end of the stick in all the places where it matters, and the compensation is not as overwhelming as on other characters. Yes Zelda has the best defensive ability in the game, but defensive ability is inherently less useful than offensive ability. In addition to that, there are plenty of characters with abusable mechanics without huge downsides, but Zelda has mediocre abilities (which are already balanced themselves) along with huge downsides. Its 5am so I won't go listing all of the examples of abuse cases I could think of, but Zelda's "abuse" case is when the opponent chooses to approach her. Even then she's just "good", much like how any character is good when the opponent does the approaching, but with punishes that feel stronger. If she could somehow force the field of battle to give her this advantage, then she would be a good character. I'm just rambling now.
Essentially what I'm trying to say is that Zelda has been carefully balanced so that all of her strengths are counteracted by weaknesses. That's great! But there are so many characters (probably all of them) that have many more strengths than weaknesses. And I know since I know my character best I will naturally notice all their weaknesses, but I also play Sheik and I don't think she has any weaknesses. Sheik is just all-around 'good' with no real problems. She doesn't have an abuse case, but she's also just plain good. Is it too much to try and get these power levels to match?
Sorry I went on a tangent. Its clearly time for sleep...
it's not forced. it's just weird out of hitstun. I like it because if you get launched, you can return to the stage quickly with a big Din's covering you the whole way down.
There's no forced fastfall or weirdness specific to being used out of hitstun. Placing it just preserves existing fall speed on startup instead of resetting it (kind of like the opposite of Marth/Roy neutral-b which reset downward movement but preserve upward). You fall the same way placing a din's out of hitstun as you would doing an aerial or just falling doing nothing. Maybe the max fall speed could be adjusted to something floatier than her normal fall speed, but it isn't an unintended property.
Hey Magus, I see lots of technical explanations, but I'm more curious about goals of or reasons for some changes. Then, I'm also curious if you feel those goals were met.
@Rizner
I didn't respond to your post directly, but if you already read those posts I don't know what more explanation I can really give you on the thoughts behind those changes that I haven't said already.
The 3.5 patch including the more straightforward bug-fixish things mentioned at the very least is a given. ICs have a gameplay crippling bug so it's not like the game is going to be able to sit until the next major release. No I can't really give a date of when that patch will be without it already being finished with how unpredictable the time required and time available can be for things, though to give an extremely safe estimate that I can't see being exceeded some time in the next few months, but probably well before then.
Ok, so this was posted in the reddit ama but not answered - basically asking what I did before but expanded upon.
From my understanding, Zelda was changed with the idea of no longer polarizing some matchups and to make her more intuitive/in line with the rest of the cast (homogenization).
Is this correct, and if not what was the goals for her as a redesigned character/what prompted the redesign?
Either way, do you feel your new Zelda idea is actually meeting the goals you set out with, primarily with aspects of her dins fire and her up-b cancel changes?
In my opinion, her design still has almost all of the traits you were trying to get rid of or change, and did not fix any of the problems she was purported to have. I think complaining will go down until Zelda's get good with her new style, then the same people will complain about the same things as before.
There's no forced fastfall or weirdness specific to being used out of hitstun. Placing it just preserves existing fall speed on startup instead of resetting it (kind of like the opposite of Marth/Roy neutral-b which reset downward movement but preserve upward). You fall the same way placing a din's out of hitstun as you would doing an aerial or just falling doing nothing. Maybe the max fall speed could be adjusted to something floatier than her normal fall speed, but it isn't an unintended property.
I'd like to talk about the nayrus love changes. I feel its pretty clear that the tweaks really affected that tool, at least the aerial version. I guess I can go along with wanting to nerf it somewhat since before it was a pretty centralizing tool of her combo game. So, with that said, I'm going to try to curb my perspective a little away from the immediate visceral feeling that one of her primary tools was destroyed. But even so, I do feel that some of her options with nayrus were removed and would like to open a discussion about if that's necessary.
So, a couple thoughts, and I'm interested in you guys' opinions on these too,
A) Aerial nayrus did a few things before, such as, helping you escape combos, get back to the ground, recover, and land cancel nayru was sort of an approach tool before. But first, here's what I mean about it as a zone breaker:
The video sites that most dash attacks are zone breakers since the character bursts forward with a hitbox quicker than most people can react in certain situations. Nayrus actually slows your approach, but the invincibility allowed you to zone break by slipping through someone's attack, if you can make a successful read.
Now regarding gtfo moves, most are quick, strong KB moves that cover much of the character's hurtboxes. Nayru's hitboxes weren't quick, and their weak multi-hits made it a poor gtfo in those regards, but the invincibility came out on frame 5, which made it a fairly quick way to slip through a followup attack and return to neutral.
Thus just by removing the invincibility, the PMDT removed 2 separate utilities: 1) zone breaking and 2) a gtfo move to fend off foes.
I feel like the PMDT wanted to tone down nayrus some but were undecided on how much, what to preserve, and how to go about doing that. People have long been complaining about nayru's invincibility (for some reason I've never understood) so I understand that change; however, I dont' feel that speeding up the hitboxes to frame 8 was the best decision to ensure aerial nayrus utility after the change. If it were up to me, I would have either sped the hitboxes to frame 5, or made nayrus give her a slight boost in forward momentum. The former would preserve more of the defensive capability(and clearly the intention of the move was to be defensive) of nayrus for fending off foes, whereas the latter would preserve more of the approach/zone breaking utility, this time by allowing it to rush in slightly quicker.
B) A reverse hitbox was added so that if the opponent is just clipped from behind it, they get send outwards instead of crossing up zelda and ending up in front.
"cleaned up hitboxes so that being hit by the edges of the last hit has reversable KB, but being hit within the outer edges still sends the way Zelda is facing"
I don't really see the issue with why this was necessary, but I don't think it matters all that much. However, I think the way the hitboxes work now is kinda buggy, and doesn't seem like it accomplishes the objective. I've hit someone plenty of times where they are in the middle of nayrus, only slightly behind her and they get sent behind zelda instead of in front. I think that if you want to make only the outer edges reversable, then have the outer hitbox behind her have the very lowest priority so that if the opponent is close enough to hit any of the inner ones, it's still the normal instead of the reversable hitbox.
There are 4 hitboxes, and all are scaled based on model size, which is controlled by the 16A68/16A6C floats (parent article scaling based on charge percentage). If return hitbox size is 2, manual 8, base size 1x, and max size 4x, that's a difference of 6 at min and 24 max. Any amount I adjust one of those hitboxes is only going to more significantly impact the larger charges was what I was saying. The explosion/non-explosion hitboxes and GFX are all linked by the parent article's model size.
Took me a little while to think of this, lol, but saying you wanted to increase the size of the smaller explosions without drastically effecting the largest ones, why not just reduce the variable that scales the sizes based on charge time so it doesn't scale so steeply?
I think what Magus is saying is that the scaling of the mine hitbox is good right now, but the explosion hitbox should scale at a different (less drastic?) rate. Since both are tied to model size I think its an either-or kind of situation.
I feel like the PMDT wanted to tone down nayrus some but were undecided on how much, what to preserve, and how to go about doing that. People have long been complaining about nayru's invincibility (for some reason I've never understood) so I understand that change; however, I dont' feel that speeding up the hitboxes to frame 8 was the best decision to ensure aerial nayrus utility after the change. If it were up to me, I would have either sped the hitboxes to frame 5, or made nayrus give her a slight boost in forward momentum. The former would preserve more of the defensive capability(and clearly the intention of the move was to be defensive) of nayrus for fending off foes, whereas the latter would preserve more of the approach/zone breaking utility, this time by allowing it to rush in slightly quicker.
Toning down Nayru's Love was probably done to negate its nigh-unbeatable combo and approaching potential.
In all seriousness, Nayru's Aerial Love was one of Zelda's lesser menacing tools and only received any mileage in the hands of the experienced. It's nerfing was most likely to negate its usage in the defensive facet of her moveset, which just so happens to be one of her only defensive options in the air. I suppose, logically, there should be no reason why a non-projectile attack would be getting past the hitboxes in the air unless it had a good reason to be there, meaning that anyone who is comboing you deserves to get the hit that would be normally negated by Nayru's invincibility. Touché to myself. I think your proposals are very interesting, especially by giving it stronger approaching properties or combo potential.
B) A reverse hitbox was added so that if the opponent is just clipped from behind it, they get send outwards instead of crossing up zelda and ending up in front. I don't really see the issue with why this was necessary, but I don't think it matters all that much.
I don't really know. This one seems like an arbitrary change to me as well. It was probably to make the move less jarring, what with it sucking you in, whipping you around, and vomitting you out in front of Zelda.
I could see that working. The issue is not having ANY of it. Like @otheusrex
said, losing invincibility, DD, AND LJ made the option so much worse than it was. It went farther than "toning it down" to the point of "crippling" it (like a fair amount of Sonic's nerfs). Also, it is just me, or were the aerial hitboxes on Nayru's shrunk? It seems to be that way, but I have no way of actually knowing. There's nothing in the changelog, but it still feels smaller.
Also, it is just me, or were the aerial hitboxes on Nayru's shrunk? It seems to be that way, but I have no way of actually knowing. There's nothing in the changelog, but it still feels smaller.
I apologize for bumping this with a new post (if anyone here happens to be bothered by that sort of thing) but as this thread seemed to be the most official one for raising concerns about glitches to be fixed. I see sheik's not carrying over the horizontal momentum of transform from zelda as a glitch; Zelda can be flying horizontally through the air all the way through the light curtain part of transform, only to stop dead in the air as soon as sheik appears. This makes for a bizarre looking and unintuitive arc, as well as diminishing the usefulness of what seems to be transform's new purpose as a recovery/escape tool.
I realize that the fact that Zelda can boost transform out of hitstun for added horizontal momentum wasn't really an intentional thing to begin with (much like love jump), but as the PMDT removed love jump, calling it a glitch, but decided to keep boost transform, even strengthening it by decreasing the fall speed during it, it seems like they're saying that boost transform is here to stay. So, in that case, might as well iron out the remaining kinks on the mechanic into something that seems whole-heartedly on purpose.
Taking a wild stab at how to keep a nice smooth continuation of the momentum Zelda had into shiek's reappearance, I'd guess you'd need to set up a variable outside of the pacs to represent Zelda's horizontal momentum so that on the Transform reappearence subaction, you can set shiek's horizontal momentum to that outside variable.
Also, I think speeding up the start-up animation of transform would do a lot to improve transform's utility, strengthening its use as an escape tool, which would partially make up for the heavy hit she took to her ability to escape with the removal of love jump, diamond dive, aerial nayrus invincibility, and increased landing lag on teleport. I doubt that transform loading time will ever be improved, and since for some reason they decided to remove Zhime's addition to shielda with the buffered aerials out of transform, transform might as well have some new tangible reward for using it aside from the purely theoretical advantage of switching the match-up mid battle.
Is the data in the frame data thread accurate? I noticed that the fair and bair's frame data are nearly identical. One thing that stood out to me is that fair's frame data claims that it sends opponents at a 361 degree angle, which isn't a change in the changelog.
Is the data in the frame data thread accurate? I noticed that the fair and bair's frame data are nearly identical. One thing that stood out to me is that fair's frame data claims that it sends opponents at a 361 degree angle, which isn't a change in the changelog.
In a nutshell: Defined as 361°, it means the higher the damage on the opponent, the higher the angle they get sent at. Additionally, when used on an aerial opponent, they are sent at a set angle
Will Zelda get a new costume?
What matchups might get more manageable?
Will she receive any significant changes?
How will the new stages and alternates affect her?
Hmmm, I'd guess Fox, Ike, and Roy are going to get small nerfs. I personally think that Marth and Mario need to be toned down a bit, but that probably won't happen.
Act out of telecancel 2 frames faster. Camera no longer tracking unlinked Din's. Those 2 things have been confirmed.
As far as speculated changes...I don't know? Something in me expects some sort of change to Nayru's, whether it be with the momentum stuff, prioritization of hitboxes, SDI modifiers, LC time...there are just SO many parts that could be adjusted, it seems logical that they'd mess with it some. I think we can also expect some change to the new Din's other than just the camera. Once again, I don't know what that'll be, but this was sort of an experimental build for Din's. Now that Magus has gotten to see how his idea worked and didn't work, I can see him tweaking hitbox sizes/strengths or a few other properties. It's only logical.
How will the new stages and alternates affect her?
IF SECRET SHINE IS LEGAL, I'M GOING TO PICK IT EVERY TIME. I loved Drac's Castle, and it was a pretty solid stage for her. Zelda tends to do well with moving platform stages (not including Norfair, because new Norfair is trash, and I miss the old one), and it has large blastzones, which tend to favor her. Yay survivability! On that note, I heard someone say something about them shrinking the blastzones some to make it more "competitive." Drac's Castle had marginally wider blastzones than DL, and a ceiling that was noticeably lower, but still 2nd highest (a bit higher than Skyworld). I don't know if that'll be confirmed, but I'd love some shorter side blastzones. Zelda doesn't need all THAT much space coming back, and super large blastzones make killing with her hard
Wow...that was long. I guess I'm really excited for 3.6!
For speculated changes, I wonder if U-smash's damage and knockback will be altered, so it can't be chained together for large damage as easily.
If Nayru's is changed, I hope it's back in the direction of 2.6b because the move went from being a great (and balanced) tool to an average tool. Aerial Nayru's is crap compared to the previous version. Opponents can fall through or out of the attack. A lot of attacks beat it now (and it can still be shield grabbed), so it's much less safe as an approach tool, which lessens the practicality of Nayru's glide. I recall Love Jumping being attributed to a glitch and the removal of Diamond Diving and it addressing that glitch while addressing global recovery nerfs, but Nayru's Glide is hardly a satisfactory replacement since it covers little vertical distance and suffers the same weaknesses as an aerial Nayru's in 3.5. Last but not least there were the strange reversibility changes. You can't reliably follow up on opponents when you don't know which direction they'll be knocked. That unreliability also weakens Zelda's ability to edgeguard or set up and edgeguard with Nayru's. Zelda has never been stupid, but some of her best moves have slowly been changed for the worse in each iteration, and I think that's to fit with global design goals, but a Zelda main must think that the public's bitter feelings toward Zelda also have an influence combined with her lack of representation at the highest level or in the PMDT. I'm not pointing any fingers, just speculating.
recall Love Jumping being attributed to a glitch and the removal of Diamond Diving and it addressing that glitch while addressing global recovery nerfs, but Nayru's Glide is hardly a satisfactory replacement since it covers little vertical distance and suffers the same weaknesses as an aerial Nayru's in 3.5.
Love Jumping and Diamond Diving were two of the many momentum bugs that came over from Brawl. To my understanding, they were left in because it was something neat that the character had. Love Jump got you out of combos, but left you high above the stage without a jump, not exactly the best position for Zelda...I have NEVER seen Love Jump effectively used as a recovery tool, but I could see it being useful in some incredibly rare situations. There are plenty of "bugs" in Smash that HELP gameplay rather than hurt it.
Most characters have something interesting happen if they use one of their special moves off an edge. Tink/Link can do a falling up-b. Mario's cape, Luigi's fireball, and Bowser's Klaw (among others) all get a momentum boost.
I know that Magus had intended to remove Love Jump and Diamond Dive entirely, but only corrected for the vertical boost, not the horizontal. That's why we still have Nayru's Glide.
I'd love to see the return of Love Jump. Mario, Link, CF, Sheik, and other characters that can juggle her have almost too easy a time once they get their hands on her. One of the "bonuses" to being a floaty was originally that it was hard to be combo'd. With a greater number of viable characters, there are plenty out there who can combo AND kill her easily. Love Jump and aerial Nayru's were her ONLY combo breakers. With Love Jump completely gone and aerial Nayru's no longer having invincibility, she is left without any combo breaking ability, an inherent flaw in a character that is supposedly "difficult to combo"
Love Jumping and Diamond Diving were two of the many momentum bugs that came over from Brawl. To my understanding, they were left in because it was something neat that the character had. Love Jump got you out of combos, but left you high above the stage without a jump, not exactly the best position for Zelda...I have NEVER seen Love Jump effectively used as a recovery tool, but I could see it being useful in some incredibly rare situations. There are plenty of "bugs" in Smash that HELP gameplay rather than hurt it.
Most characters have something interesting happen if they use one of their special moves off an edge. Tink/Link can do a falling up-b. Mario's cape, Luigi's fireball, and Bowser's Klaw (among others) all get a momentum boost.
I know that Magus had intended to remove Love Jump and Diamond Dive entirely, but only corrected for the vertical boost, not the horizontal. That's why we still have Nayru's Glide.
I'd love to see the return of Love Jump. Mario, Link, CF, Sheik, and other characters that can juggle her have almost too easy a time once they get their hands on her. One of the "bonuses" to being a floaty was originally that it was hard to be combo'd. With a greater number of viable characters, there are plenty out there who can combo AND kill her easily. Love Jump and aerial Nayru's were her ONLY combo breakers. With Love Jump completely gone and aerial Nayru's no longer having invincibility, she is left without any combo breaking ability, an inherent flaw in a character that is supposedly "difficult to combo"
Based on the bold parts, wouldn't the return of Diamond Diving be better if Nayru's regained its aerial invincibility? Diamond Dive helped one return to the stage faster and break combos without putting you in a disadvantageous position without your double jump.
Anyways, you brought up a great point about a new inherent flaw that they gave her. 3.5 weakened her approach options, ability to dictate the opponent's movement or pace of the match, her combo breaking ability, and more. She received some seemingly arbitrary buffs that don't make up for the losses that push her viability closer to Brawl levels. I don't see how the top Zelda players would fare versus the top players of Melee top tiers with her reduced options. M2K is a great player that beat Zhime twice in Zhime's grand debut at Pound V.5. Considering how Fox, Sheik, and Marth have remained about the same while Zelda was scaled down, I don't see how Zhime or anyone would win those matchups at the highest level.
Adding Diamond Diving back in would be fantastic because it gave her an amazing option to get out of juggles, something she really struggles with in her current iteration. It also takes some finesse to know when to apply it, so it's not exactly a "get out of jail free" card, especially with the removal of invincibility on aerial Nayru. I'm also looking forward to teleport fixes; ideally, those 2 extra frames of endlag will be removed and it will stop getting stuck in walls. A more lenient timing window on shortens to platforms would also be great, because as it stands now, it's almost impossible to shorten up to a platform with any degree of consistency. I used to have no issues with this when Teledashing was still a thing, but it's so risky now because missing means getting stuck in a lot of endlag.
Aesthetically, I'm banking on new OoT color options. I'm immediately switching my color to OoT white or blue if these are added.
Hyrule Warriors Zelda would be so cool. I don't know if I would want to use OoT blue Zelda since it was such a light shade in Melee. I would totally use a blue Hyrule Warriors Zelda though.
In 3.6 I really just want the camera scaling removed from Din's Fire. It gives me a headache playing on Dreamland with a small CRT and EVERYONE ******* about it. I'd like it if teleport didn't get stuck on walls but they didn't change that in 3.5 so I doubt it will ever change. After playing with @otheusrex
's Farore's Wind I would actually like to see something like that added into the main build. I'm sure a lot of other Zelda's would disagree with me, saying it would be a major recovery nerf (our teleports would be telegraphed), but it really helps with practicing teleports and knowing where you are.
What does everyone else want for 3.6? Critical Heels returned? A better dair hit box? Diamond Diving? Smash 4's teleport? Melee's nair?
