HisoutenSoku

So, Tasogare Frontier (aka Tasofro) released an expansion to Scarlet Weather Rhapsody. The game is literally an add-on. They added five characters (spoilers lol yeah right: Cirno, Hong, Sanae, Suwako, and Utsuho), revamped the HUD and menu, and added some music.

Last time around when I played SWR it was mostly with Alice and I had a poor opinion of it. This time around I decided to play as many different characters as I could stomach and treat it as if I was starting from scratch. However, it turns out there's still some good/bad/ugly, just like last time. For example, on the ugly side they never really fixed Reisen and Sakuya's bogus sprites, and Sanae's sprites also look rather awkward.

I will say that they did fix a handful of issues that I had with the game (such as added pushback on wrongblock), but then they went and added another handful of other issues. So, in a way it kind of balances out. Except that I just don't like the core gameplay at all in the first place, which hasn't changed.

It occurred to me that throwing bullets puts you at a disadvantage unless it's airtight. If the opponent grazes and you cancel the bullet with a movement then the opponent will invariably be able to attack before you can. Be it a; high jump, back dash, airdash, or fly to cancel the bullets on the ground or in the air it will put you at disadvantage. They will hit you before you can even attack, and it's because HJ cancels and airdash cancels are so incredibly slow. That means whenever you throw a bullet the only way you can maintain advantage is to cancel into a special move and pray it works, preferably a melee special move.

Your other option is to throw extremely slow moving bullets from a safe distance so that you can move into the bullets and use it for cover, but this accomplishes nothing due to the extremely huge stage in SWR. Wherever you throw bullets and follow them to, the opponent can just avoid your setup entirely. So what if you can cover the entire bottom half of the stage when the opponent can just fly to the top and opposite side of the stage, completely avoiding you and your bullets?

Additionally, there's really very little penalty or risk in airblocking. It's extremely hard to guardbreak some one with a melee anti-air when there's 8-way flight. A good cautious player will never be afraid of melee anti-airs.

Speaking of anti-airs, the best anti-airs in the game are all bullet moves. The majority of melee anti-airs suck horrifically or are prone to whiffing due to 8-way flight or just flat out losing to fast air bullet attacks. I mean literally, if some one on the opposite side of the screen does hj9 d[6] and flies across the screen at you then you better hope to have a bullet anti-air because there's practically no point trying to melee anti-air them, even when you see them coming at you from two miles away and know exactly what they are doing and where they are going. Because as soon as they are within range of your melee anti-air they can 8-way flight with the second airdash in any direction, or throw a bullet right in your face to beat your melee anti-airs.

However, that brings us back to the previous issue of throwing bullets putting you at disadvantage. If the opponent predicts that you'll attempt a bullet anti-air and just grazes through it then yeah you're at disadvantage. Once again the best option is to throw a projectile then cancel into a melee special move, which characters like Alice don't even have without alt.skill cards. And even then it's no guarantee since the melee special could whiff if they fly right over you or backwards out of range.

The best characters in the game have really fast movement and good melee. In SWR1 it was Youmu, Aya, Marisa, etc. I can see Cirno and Hong being added to that list simply by virtue of good movement and melee alone. They don't even need bullets or supers, but certainly having ridiculously powerful supers helps (which they do have).

In retrospect, it really feels like they made the entire rest of the cast feel more sluggish and slow in HisoutenSoku. Playing Iku Nagae or Komachi for a few rounds then switching to any of the aforementioned characters makes me cringe, they and so many others in the cast simply feel like molasses by comparison.

That being said, the new weather types aren't that big of a deal really, but the existing weather types are now either the same or even more problematic than they were before. Mountain Vapor randomizing cards in addition to hiding them is ridiculous. River Mist is even more obnoxious since it's random and stronger. And my personal favorite Spring Haze now lets you graze both melee and projectile, which means you can just wavedash graze indefinitely for the entire duration of the weather and you won't be hit by anything. And yeah, Typhoon is still just like it was before, superarmor for all. Oh and if you can't read Japanese; that weather in the image is Typhoon, so you probably wouldn't know both players had super armor until some one got hit.

The new system cards are ridiculous. The most stupid ones are the ones that grant super armor or absolute invulnerability for a duration. The invulnerability cards requirements are pretty steep, but if you get it off then the opponent can't really do anything at all except run away. And run away is exactly what most players do, further illustrating how easy it is to just run and turtle in the air on such a huge stage with the flight ability. I hear some players complaining about the low damage for certain supers, but casting 3 cards for invulnerability only to get maybe 700 damage out of it if you're lucky is quite a waste. I think the other cards are for the most part benign and only marginally useful, but the magic potion card that refills your spirit almost instantly is just one more thing that makes having broken spirit orbs kind of meaningless; since yeah it not only fills empty orbs but also broken orbs, and you can act almost immediately after casting it. So even if you manage to crush a bunch of orbs, a 1-stock card that puts them all back instantly is available that reverses all your efforts. Which also makes the penalty for Border Escape (guard cancel dash/jump) even less important.

In closing, I'm not saying it's a bad game. But it's definitely not a serious fighting game. Tasofro has clearly stated in their blog that they designed it to be more of a casual game, and it shows plain as day. That being said, I'm personally really not into multi-playing in casual games. If I play against other people there is going to be some competitiveness inevitably, and trying to force any amount of seriousness into a casual game will just end in total failure.

---

- Copyright © Xenozip.

Situational

The answer to every IaMP related question is: "Depends."

The reason is because there are quite a lot of situational variables to take into consideration for any given scenario, so it's impossible for any given answer to be right all the time. It's just one of those qualities of IaMP that is both rather frustrating and yet can also be endearing in retrospect.

For example, in most fighting games you could say: "If the opponent jumps at you, do c.HP or DP as an anti-air." and it would apply in almost all situations where the opponent jumps at you. It's a good general rule of thumb. Sometimes this even leads to players not jumping at all unless it's after a knockdown, because there's never a right time to jump except on okizeme.

But in IaMP you can't say that you should always anti-air the same way in every situation, it depends on many different things. Lots of things can effect your method of anti-air:
- If the opponent has no bullets in front of him and has used up both airdashes, either a bullet or melee anti-air works fine.
- If the opponent has bullets in front of him and no airdashes you can anti-air with bullets.
- If the opponent has both airdashes saved up and no bullets in front of them you can use a melee anti-air at the right distance.
- If the opponent has both airdashes saved up and bullets in front of them you your options are to use bullets to force them to use their airdash and negate their bullets, or to attempt to dash under them to graze their bullets and force them to airdash toward you.

You would also think that you should always go for a tech-trap setup in order to capitalize against a bad air tech. Or to never air tech against a good tech-trap setup. But even these have conditions. You might not want to go for the tech-trap setup if it would cost too much spirit, self draining you as a result, especially if you know your opponent respects the risk of air teching and you assume that the opponent will not tech. Some characters would also shy away from trying to tech-trap Youmu, Yukari, and Remilia due to those character's unique properties. Likewise, you may choose to air tech and risk taking extra damage in order to deny your opponent bomb stocks, because you won't be landing on the ground so you can prevent them from regaining any bombs they used. You also avoid having to deal with a meaty okizeme rushdown that would put you back into the corner.

I myself am guilty of getting hit by the same thing in the same situation repeatedly, without realizing it or fixing it. Though in my defense I think I'm like most players -- I need time to think about it before I can adjust my game. So don't worry, if you find yourself getting repeatedly lamed by something or other, you're not alone.

If you find yourself beating your head against the proverbial brick wall, you may want to take a moment to consider what all your options are and if you've fully explored each one. What do you think would be a good word to describe; trying the same thing over and over, expecting different results?

And if you do find an answer, and then that same answer fails the next time, just remember that situations can change frequently in IaMP.

- Bellreisa says: "There's never an always in IaMP."

---

- Copyright © Xenozip.

The Brick Wall

Being that this blog was meant as a personal journal, I feel I should probably make a personal post about the game. The blog is old enough now, I think.

I've likened the learning curve in IaMP to that of a brick wall for years now. I've continued to feel that way even to this day. To elaborate: each brick represents a bad habit or unlearned "function" that you'll beat your head against until you break it. These "bad" habits may come to us naturally/intuitively or be formed from other games. And these "functions" are usually the things that apply to IaMP's game-flow that we may not automatically consider or recognize, but they tend to bite us on the ass.

Simple things like: Attacking first doesn't always win, height superiority has it's advantages and disadvantages, not beating bullets with melee or visa versa, moving backwards can be stronger than moving forwards, using cover for advantage, and how the mixup and pressure games actually work. They sound simple, but if you don't "get it" it can be a real chore to learn it during a round or even multiple sets of games.

Some bricks are harder to break than others. And, all that head-banging is bound to cause some headaches. Indeed, rage and frustration is really rampant in the IaMP scene once you get to a certain skill level and continue to move upwards. This is generally caused by getting repeatedly hurt by something, realizing that something is wrong, and not understanding/recognizing what's wrong or how to correct it. That's when the head-banging begins.

IMO, you must first isolate a single brick among the wall of other bricks and then forcibly break it by grinding it out. If you're not careful and attempt to break too many foundation bricks at once then the whole wall might come crashing down on you. I know it sounds like a silly analogy, but it's actually very real and has happened to others (we've witnessed it first-hand numerous times). If you allow yourself to get too frustrated while banging your head against the whole wall all at once it'll probably end up being much too overwhelming. You'll crack before any brick even starts to. You can't expect to pick it up and be automatically good at it or level up very rapidly. All you can really hope to do is be patient and observant.

The best players in the IRC channel have been around for years, and they were trash at the game when they first started (we have recorded evidence of this). It took a long long time to get good, and the same applies to everyone else.

Know what? I think that's a good thing. A very good thing. If you consider the opposite, it'd be quite a great tragedy to have some newb pick up a game and then suddenly be playing as well as the top 3 players outside of Japan. That would suggest the game has very little to offer in terms of skill development and depth. But I say IaMP holds a vast amount of areas that we can develop and is indeed quite deep. And the same should hold true for any good game, I think.

Unfortunately though, the scene tends to be cruel to newcomers at times. When first starting out; there can seem like such an enormous gap between yourself and the best players around. Which can be even more menacing if you happen to also pick the worst character in the game.

Indeed, this was my experience. I first started playing with Hong Meirin, the inarguably bottom tier character -- and my first real opponent was Bellreisa, one of the top 3 players outside of Japan who is probably very capable of doing well in tournaments in Japan. And that kind of noob hazing is actually a bit common in the scene, I wasn't the first or the last to get demolished on a debut.

Fortunately, I'm the type that gets hungry to level up after a beating. Sure, I freaked out then, but my rage only fueled my interest and desire to learn. I thought to myself, if I could receive that much of a beating with my years of fighting game experience under my belt; then it stands to reason that the game must have a lot to learn in it, and I wanted to learn it, even if that meant more beatings.

Though, I can safely say that at the time I was in great danger of quitting because I was very stubbornly attached to Hong Meirin. Had I actually stuck with her exclusively, I think I can safely say I'd have quit long ago. It's not necessarily the character's fault or the game's fault per se. Actually, one of the fundamental things that I believe in regards to IaMP is experiencing a variety of characters. Had I stuck to just her, I probably wouldn't have learned anything. Be it due to a very narrow perspective of the game, or just due to my overall incompatibility with the character itself. It wasn't until I really tried experiencing other characters that I was able to learn more about the game, and able to realize that I actually like other characters more and at the same time realize I didn't actually like Hong Meirin at all. And that's when the brick breaking process actually really began for me.

Some of us in the scene have also noticed other "narrow perspectives" and some pattern habits that can be associated with other characters. A common one that often arises is the trends we see prevalent among the Youmu players, such as reliance on naked air attacks and tactless brute force. Further leading me to believe that character variety is a window out of this form of tunnel vision.

To put things in a little perspective, let us examine this Alice flowchart made by Bellreisa (click thumbnail for larger view). This flowchart isn't a joke, like the SF4 Ken flowcharts you see floating around the 'net. No, this is actually very much what goes on in the mind of Bellreisa while he's playing Alice, and probably other Alice players like TMN and such. The thing is, that at each junction, you need to know what all the optimal available options are to you. If you don't, guess what, that's a brick that's going to hit you when you ram face first into it. In order to fill out that chart in your head and put it all into muscle memory, that's a lot of bricks you gotta run into and break along the way before you feel competent and in control of each situation.

Fortunately not every character is as complicated as Alice. It's possible that she is the most technical and complicated character in the game. But, it should give you some idea of what you'd be looking to accomplish if you were to pick up the game years after it's initial release, main Alice, and stick with Alice exclusively while playing against players with years more of experience. It's going to be a long and windy road. It's only slightly less windy and less full of potholes if you pick another character, and IMO you can make the ride smoother by choosing to learn with multiple characters rather than maining just one character exclusively.


- Bellreisa says: "I'm busy".

---

- Copyright © Xenozip.

Commentated Matches

Bellreisa has been commentating match videos for the Japanese tournament series called "Bullet Action" and putting them on youtube. You can see the playlists for the videos here:

- Bullet Action 3 w/Commentary
- Bullet Action the 5th Commentary

I strongly recommend watching these videos regardless of your interest level with the game. It's not only entertaining, but they can also help give some insight into what's actually happening during a match.

Bellreisa mentions when a player has done something they shouldn't have, or missed an opportunity to do something they should have. Not only that, he also mentions why the players sometimes intentionally don't do certain things. I feel it can be rather educating to those who would otherwise overlook the things that he mentions.

Hearing his commentary makes me wish all tournament matches were commentated on in a similar fashion. It also makes me realize parts of why the Japanese regularly do match commentary. Which also makes me sad that I can't understand Japanese commentary.

At any rate, props to Bellreisa for doing so. I feel it was a good thing for him to do, and I feel he did a great job.

- Bellreisa says: "Bad bomb!".

---

- Copyright © Xenozip.

Scarlet Weather Rhapsody

On May 25th, 2008 we finally got our hands on the sequel game to IaMP:

TH 10.5 - Scarlet Weather Rhapsody, Touhou Hisouten.



My blog here is intended specifically for IaMP related journal entries though, so this is the only post where I'll be talking about SWR (Hisouten). If I decide I want to start making journal entries for my experiences with SWR in the future I'll probably start another blog.

So, how is it?;

Very mixed reactions to it so far. The reason is because they did so many things wrong, and so many things right. But you really have to give them credit for making the game independently. It is, after all, a doujin game. Meaning a very small group of people that were not being funded or working for any company, they got together and made it happen.

So far, it's tempting to want to play and enjoy the game, and I really do want to like it. But rather than picking apart everything and being my usual verbose and bombastic self, I'll just give a run down on the good, the bad, and the ugly -- and try to keep subjective opinions to a minimum.

Just keep in mind that this game is a doujin (independently made game).

The Good:

Netplay

Thank you Tasofro, you almost got it completely right. The game comes with a built-in netplay mode which is totally acceptable by community standards. It even comes with a spectate mode and auto-save-replay function.

Buffering

The new buffering system in SWR is actually very user friendly. Jump cancels can be done by simply holding an upwards direction, bullets can be backdash canceled, and special move input buffer windows are rather large and lenient.

Combos

No longer do some characters have combos for only some specific situations. In SWR the whole cast has a combo for every situation.

Graze Windows

The graze duration for forward dashing was increased to include the entire dash and dash cool, so transitioning from a dash recovery into another graze action is seamless.

More Characters

Of course it's a good thing. Five new characters to the original IaMP roster (Aya, Komachi, Reisen, Iku, and Tenshi). Two of the new characters being entirely new to the Touhou series (the other three being characters that appeared in the danmaku shooter games, but not in IaMP).

Tons of new Specials and Spellcards

I mave to admit, I like the fluff. It adds variety.

Bombs, Hangeki, and Border Escape

Border Escape is a purely defensive guard cancel and not a reversal. Hangeki is a guard reversal but costs a spellcard, it's slow, and reward with absolutely nothing other than a knock down. Bombs are IaMP-style N-Bombs that cost a card. I like them all.

No Hong Meirin

Knock on wood, the low tier joke character may make a return in future patches to the game, since they have her sprite set from IaMP anyway. But I rather her be gone than a joke/waste/handicap.

The Bad:

Missing Features

Netplay's spectating is kind of hit or miss, there's no way to check on a game, no control over fast forward (which is poorly done), and it sometimes doesn't work mysteriously. There's also no way to manually control delay settings in netplay, it's somehow done automatically. And there's also no lobby or matchmaking service available by default, which is probably asking too much, but players will have to find each other on their own.

Missing Characters

Yeah, I like that Hong's out. But it's also a bad thing because it's a waste to have the sprite set and not use them. We lose content because they axed her.

Blinding Stages

Three of the stages in IaMP are hard-banned by the community because they are so bright that they mask a lot of bullets from a few characters. In SWR, not only did those banned stages return, but there are new stages that are far worse. The Hakugyokurou stage, which is Yuyuko's new stage, is covered in white snow, making it impossible to see most transparent projectiles.

Problematic Weather Effects

The theme of having weather is kind of borderline. But the effects caused by some weather types are just so disruptive to natural game flow that both the English speaking community and Japanese community have agreed to use a hack that removes weather effects. Some of these effects include (but not limited to): an instant crush when blocking anything incorrectly, super armor mode that removes the ability to block anything, a weather that removes the ability to do normal melee and instantly recharges your spirit, a weather that damages your super meter when you take damage or block, a weather that rebuilds your life when you inflict damage, and a weather that pushes opponents away from each other. Each of these are extremely disruptive. And to make matters worse, the weather is chosen at random.

Card Deck System

I have mixed feelings about it. It's partly random. Each card has a 5% to 20% chance of appearing when you earn a card in-game, depending on how many you put in your deck (1-4 cards per type / 20 total cards). The fact that the cards are earned randomly during a match based on these numbers is what I find disruptive.

Skill Cards

In theory this is cool, but the execution is not to my liking. Some characters effectiveness is vastly increased once a skillcard is activated, you don't get them at the start of the match. You not only have to equip the card in your deck, then acquire the card randomly during a match, but also cast the card, just to gain the benefits of an alternate special move that overwrites an old one.

Wrong Block

There isn't much penalty for blocking wrong. In fact, there isn't much reason to block high. When you block a melee the incorrect direction (high/low) the normal amount of pushback caused is doubled or sometimes tripled depending on attack strength. The penalty for it is minor: some spirit damage that regenerates quickly, and some very minor added block stun.

Slow Crushes

There isn't much reason to use guard crushes now. They can't be canceled into bullets or specials. When blocked right they are highly punishable. Blocked wrong they don't reward with much of anything unless you have a level-5 super stocked. They are also too slow to be used as frametrap staggers.

Flight and too much Space

The size of the screen can be panned to include the entire width of the stage. Characters can fly to the top of the screen, which is almost five character lengths high (Alice stacked on top of herself four and 3/4ths upwards). Flight is omni-directional, allowing you to escape by flying upwards or downwards or diagonally, etc.

Controls

In most cases, flight is slower than airdashing. And flight uses up an airdash from your max. And because flight is D, holding D and upwards from the ground causes first a superjump, then quickly cancels the superjump into a flight. So for a natural superjump you need to use a doubletap. Likewise for airdashing, you need a double tap. This is problematic because using D for air movement is more ideal, but using flight is less ideal than airdashing or superjumping in a lot of situations.

Forced Damage

A few characters have ways of forcing damage with attack strings, by crushing an orb with bullets and specials, off just about any blocked move. Anytime, anywhere, if you block something then there's absolutely nothing you can do about it. The only way to avoid damge is with a Border Escape or Hangeki (guard cancel and guard reversal respectively) which can be baited and punished. The difference between a normal scenario in other games, and the one here in SWR, is that here you're not skillfully defending yourself from damage, you're forced to eat damage or risk a punishable defensive move to avoid eating the forced damage.

Enormous Tech Roll Lengths and No Suki

In SWR you can wakeup graze by simply holding D and a direction, you no longer have to reversal graze. Combine that with the fact that tech rolls cover half the screen, setting up tech traps that cover two or more tech directions is highly improbable, maybe impossible.

Proration and Punishability

Although they are disruptive, I consider these minor because you can work around them, and they could even be adjusted/fixed in future patches. But the proration for combos and the punishable windows on specials/supers is ridiculously haphazard in design, IMO.

The Ugly:

Reisen

Her sprite is painful to see. She is so short in stature, and she's waring what looks like a school girl outfit. It's so terrible and sad.

New Stages

While I do like the busted shrine, I don't like any of the other new stages. They are all really bland, boring, and ugly. With one possible exception being Aya's stage, maybe.

Missing Stages

It's cool that they brought back some stages and then added more. But where are the missing ones? I'll never understand why developers remove content between sequels instead of just adding and adding. I don't even mind that they didn't adjust the rehashed stages much/at all, but if you were going to rehash some then why not rehash them all plus add new ones?

New Music

I really like two songs from the new soundtrack. The rest of them I hate. They are bland and similar to each other. Oh, how much I loathe the accordion remix of Yukari's "Evening Star" theme.

The New Sounds

Maybe a lot of the new sounds are kind of cool, but then you get to Yukari and it's just painful. What were they thinking.

Single Player Mode

It's really as unfun as always. I wasn't expecting them to add any new features, but a survival mode or some sort of alternative game-play mode like "time attack" would have been more interesting than the standard story/arcade modes, which are far too simple and linear.

Language Barrier

It is a Japanese game, so by all means it should be in Japanese. But, we're on the short end of the stick for this one, because unless you can read Japanese you're going to have trouble with the game's menus, card descriptions, storymode dialog, and most importantly the weather.

Bottom Line;

All in all, I want this game to be something I can enjoy, and I will keep hope that some future patches will fix some of the bad points to it. In the mean time I think I'll continue to play both IaMP and SWR.

---

- Copyright © Xenozip.

Drip drip

Hehe, don't ask.

- Copyright © Xenozip.

Baiting and Provoking

In all Fighters there's many ways to set up traps using the archetypical hook-line-sinker to bait the opponent into a trap or trick. Due to the movement system in IaMP there tends to be instances that are somewhat more common due to seemingly intuitive reactions for many situations. A particularly common occurrence among newer players is to pass through bullets by forward dashing and then highjumping into an airborne opponent, such as in the above image. The route of this movement is outlined with the yellow arrow. Perhaps because using D6 then pushing 9 will transition you from a dash into a highjump very seamlessly, we sometimes do this without thinking.

-
However, the bait was Yukari herself, floating around in the air where she appears vulnerable. The trap occurs once Yukari air dash cancels her bullets and uses height, range, and speed to her advantage. The result of this exchange is often like what you see in the above image, with Yukari at a height and range advantage. Here, Yukari is free to flop Youmu right in the face, causing a counterhit ground bounce which leads to a massively damaging combo. The speed advantage Yukari has comes from the fact that she is able to attack immediately after the airdash startup ends, whereas Youmu can't attack until first passing through all the bullets, which doesn't happen until some time after Yukari has already started to attack.

Though, this certainly isn't limited to just Yukari, many characters have bullets that they can cancel in the air very quickly, such as; Reimu, Marisa, Sakuya, and Patchouli.


It's also all too tempting to try and do a graze attack in a lot of situations. But graze attacks are often laggy, either on startup or recovery. Therefor, we can attempt to take advantage of this by provoking our opponent into doing it when we want them to like in the above image, then punishing it.



Since we have two airdashes, we can not only avoid these attacks but also punish them. The Sakuya player in the above image first used an air backdash to avoid Patchouli's D3B anti-air, then after the attack had whiffed Sakuya used the second airdash to kick Patchouli in the face, as seen above.

Not all baiting occurs in the air, though. I'm not really sure where the bad habit of jumping at your opponent comes from -- within ourselves -- but it's rather intrinsic to the Fighting game genre in general to try it, even when we know it will probably fail. In IaMP it's a bit easier to goad the opponent into doing such a thing, partly because anti-airs in IaMP aren't quite as consistent and guaranteed as they are in other games (like CvS2 lol), and also partly because of the way grazing works.


Firing off some bullets on the ground that recover quickly, then covering the nearby horizontal area with ground melee, allows us to set up yet another trap. Our opponent knows that we can cover the ground with some large attacks before they can move close enough to attack, so it can be rather enticing to try and jump over them and hit from above.


But this isn't an uncounterable action. As we see in the case of Yukari, we can jump back and attack from far away which takes advantage of range. We can HJ8 and attack from above which takes advantage of height, since HJ8 moves vertically higher and faster than an opponent's diagonal jump does. We can also anti-air with a ground normal provided we have a good one. And we can use a graze attack or special move with graze frames, such as Yukari's 623B.

The correct action the opponent should take is not to fall for these obvious traps in the first place. But rather, taking evasive actions and then setting up their own array of bullet cover is really ideal, as outlined in my Moving Backwards to Win post.

So what we end up with is a game of actions, counter actions, and countering counter-actions. Thus, nothing we do is really just free and simple, but rather we have to work for our rewards.

---

- Copyright © Xenozip.

Reversals, Gaps, and D

Reversals have been around the Fighting game genre for ages, but they often come with their own set of rules or quirks depending on the game. Generally, it's the act of performing a special/super move when getting up off the ground (after being knocked down), known as "wakeup" reversals.

First, in quite a few games when you perform a successful wakeup reversal the HUD will notify you with a "reversal" message, and this doesn't occur in IaMP at all, so in some cases it's hard to know if you did it right or not. Second, IaMP does indeed have it's own set of quirks regarding wakeup grazing, backwards reversals, and frame "Suki" which means frame "gap".

The 1F Suki;



In IaMP there's a special mechanic called Suki (or Gap) that refers to a situation where a bullet move connects 1F (one frame) after blockstun, hitstun, or wakeup invulnerability ends. The easiest and most common example of this would be to use a bullet on a knocked down opponent just as they get up, such as Patchouli's f.2B bubble move as seen above.

If you simply hold D and a direction in this situation you won't graze the bullet. You will either be hit by the bullet or block it, depending on if you were holding a backwards direction or not (1, 4, or 7). The reason is because holding D and a direction will not trigger an instantaneous dash/highjump movement, but instead there's 1F before the action begins. I like to refer to this as a "wakeup dash" attempt, which I refer to independently of a "Reversal-Graze" attempt.



For dashing and highjumping, IaMP treats a reversal input different from a held input. It is technically possible to graze a bullet on wakeup using the D button, however this requires you to push the D button -- or to push a direction while holding the D button -- on the exact one single frame before you're hit, known as a Just-Frame. Doing this would be considered a Reversal-Graze since it requires "reversal timing".

Directional Input Reversals and Cancels;

As outlined above, the D button requires a 1F timing to use as a reversal, however there is another input method to perform a graze and that's with a double-tap. Such as dashing with 6-6, or backdashing with 4-4, or highjumping with 2 then 8 or 1 then 7, and so on.

This method of input is a lot more lenient in regards to reversals, among other things. Using this method, IaMP will allow you to input the last part of the input a few frames earlier than is necessary and it will still perform the action on the first possible frame. In other words, unlike the D button, this method doesn't require 1F timing. Instead there's a larger window for when you can push the button that will result in success.

The rundown in laymen's terms: If you just hold D and a direction you will not graze, you'll get hit. If you tap D at the right time it requires a 1F (perfect timing) input. And if you double tap your directions there's a 5F input buffer window, resulting in a "reversal graze".



But this doesn't apply only to reversals, this also applies to bullet cancels. For example, using D to cancel Remilia's 236C into a highjump on the very first frame that the move is cancellable, you must once again input the D button on the exact right frame in order to cancel "as soon as possible". In other words, you can't input the D button any time before the move is cancelable, you can only input D on the same frame that it's cancelable or any time after that frame. But using your down-then-up input method, such as 2 then 7, the timing again becomes more lenient. It's actually 5F (five frame) input window, if you're wondering. That means if you tap down then push upwards four frames before the cancel window it will cancel into a highjump on the fifth frame (the first cancelable frame).

Indeed, upon inspecting many Japanese matches and examining their inputs with an input viewer, the majority of the best Japanese players (AKA. the "top players") use directional taps to cancel bullets and for Reversal-Graze, but they use the D button for pretty much every other kind of movement. In other words, they only use directional taps for HJC and Reversal-Graze, and D for everything else.

Backwards Reversals;

To demonstrate the concept of what happens when you get up facing the wrong way, let's look at a video:



High quality youtube version: Reversal Graze (1F Graze) from behind

As we see in the above video, another peculiar quirk occurs with reversals which I call "Backwards Reversals". This happens when you are knocked down and the opponent ends up on the other side of you or "behind" you. Thus, when getting up off the ground your character is facing the "wrong way", meaning away from the opponent.

In this situation a couple of peculiarities occur. First, reversals are done by inputting the directional buttons the way that your character is facing, and not "toward" the opponent. Second, reversals turn you around automatically to face your opponent, including Reversal-Grazes.



To use an example we will say that Marisa is knocked down, and then uses a ground tech roll forward and ends up on the other side of the opponent. Now, in order to do a reversal uppercut (623A) you would think that you'd need to push the first direction toward the opponent, so in the above image that would be Right (towards Reimu). But this is not the case for IaMP. As outlined above, you instead need to push the direction you were facing when you were knocked down, so like in the above image you would push Left first (away from Reimu). It kind of feels like you're inputting the DP motion "backwards", but in reality you're always inputting it "forwards" relative to the direction you're facing and not relative to the opponent's position.

You'll know if you did it right because Marisa will perform the Miasma Sweep uppercut toward the opponent once she gets up off the ground, but failing it will result in either no attack or possibly her 214A broom ride move instead of the Miasma Sweep.

Now, this also applies to Reversal-Grazing as we saw in the video posted above. When you double tap the direction, it's not relative to where the opponent is, but rather it's dependent on what direction you were facing in the first place. So after crossing the opponent, while you're facing the "wrong way", double tapping toward the opponent results in a backdash instead of a forward dash.

One last peculiarity that occurs with this is Highjumping. Normally you are able to Reversal-Graze with a highjump either forwards or backwards. But after crossing the opponent up and performing a Reversal-Highjump you will always either jump vertically or towards the opponent. Both HJ7 and HJ9 result in you jumping toward the opponent rather than away, regardless of how you input it.

Seems odd indeed, but this is actually very beneficial. The reason is because it prevents crossup shenanigans from countering reversals. You see, in other games if the opponent were to use an airdash or two airdashes to cross over your fallen body, with the intention of screwing up your reversal inputs, this would actually succeed because inputs are relative to their position. In IaMP though, you need not worry about that because the Reversal inputs don't change when your opponent changes sides at all. Attack reversals also won't point away from the opponent when you succeed, but rather the only time such a reversal ends up going the wrong way is when the attack begins and the opponent crosses you up after the attack has already started.

---

- Copyright © Xenozip.

IaMP Tutorial Video

High quality version: Immaterial and Missing Power Tutorial

So, here we have it. A collaboration between Bellreisa and I, with assistance from Mauve. Bellreisa was in charge of the narration and video production, while I was in charge of the video editing and direction, and all three of us were responsible for recording the material and for general collaboration. The video should be self explanatory, so I'll talk about things surrounding the video instead.

The concept of a narrated demonstration tutorial was something I've been wanting to do for the longest time. I don't even really recall exactly how long ago it was, but it was when I first saw the Street Fighter 3, Third Strike Anniversary Edition DVD that included a video tutorial with both verbal and written explanations, oh so long ago. You can see a sample segment of the tutorial regarding Gouki here.

Later, I bumped into David Sirlin's Super Street Fighter 2 Turbo tutorial which was a representation of exactly what I had in mind; a video that demonstrated a concept both verbally and visually, and then showed it to you again visually but in slow motion and with graphical visual aid. You can see a sample segment of the tutorial regarding beginner techniques here.

Then even later I stumbled on Buktooth's Capcom versus SNK 2 video "tutorials" which were really simply just a recording of him explaining different character's core gameplay to some one else in the room. Sounds simple enough, however when I think about it I realize this is exactly what an ideal kind of tutorial would be; the act of explaining something to some one as though you were talking to them in person (just minus the actual interaction). You should be able to see a sample segment of this stuff regarding K-groove Rock here and part 2 here.

Certainly, all of these were huge influences for me in different ways. With these things combined I had become somewhat obsessed with the idea, so I was certainly poking around and fishing for opportunities to give it a shot myself. Finally, around December 28th of 2007, a couple days before my birthday, the concept of an audio+video narrated tutorial for Immaterial and Missing Power started to spring to life over a late night IRC chat conversation.

Quite a jump from late December '07 to early May '08, roughly four months plus one week, but there was a lot of deadspace in between dates where we just talked about the video rather than did anything. Or times when one of us was procrastinating and just doing other things. In reality the video didn't take nearly that long. I was able to complete about 3 whole sections of the video per single day once we had all the material recorded, and worked on it on separate days. The only section that took one whole day on it's own was the first section, because I needed to appropriate the footage that you see in the intro and outro, and also create the ghetto button input screens and other such panels. I also spent a few days deciding on a format and whatnot where I didn't work on the video at all, but rather I just toyed around with static sample junk in Premiere.

Now, just to clarify, when I say that Bellreisa was the Producer and I was the Director, I mean it quite literally. I decided back when we first talked about it that Bellreisa would have full creative control over the script for the narration, and also control over what visual content would be used to demonstrate what was being said. The only creative control I had was over the concept design of the presentation itself, and whatever I was able to do with video editing in Premiere. So in a way you could say that it is really his video, and all I did was encourage the idea and then presented the content that was given to me.

Anyway, that's the brunt of the backstory on the video. No need to get into further nitty gritty details or anything. In future posts I'll try to get back to talking about the game itself.

---

- Copyright © Xenozip.

Reading Yomi

Yomi is a Japanese word that literally translates to "read" or "reading". So the title of this post sorta means "Reading Reads". This will be a followup post to my previous post regarding Frames and Numbers.

What it refers to in Fighting Games is essentially the act of predicting (reading) the opponent based on educated guesses and weighted values. In some Fighting situations, there are some things that are simply too fast to react to, and in those cases you must use anticipation in order to take advantage of a situation. So, we must make assumptions about what will happen, done in advance. The difference between a reaction and an anticipation is simple; a reaction is done after you've already seen what is happening and "know", while an anticipation is done before you know what is going to happen and "assume" (or guess). Reactions can only fail if your execution fails or your reaction is slow, while anticipation can fail if you guess wrong.

There are different facets to Yomi which is referred to as Yomi layers, and there's three of these layers (or "levels"). If we label these layers as Layer-A, Layer-B, and Layer-C, then we can say that A < B < C < A (like math, A is less than B, and so on). While two layers of equal value nullify or cancel each other out, so A versus A = neutral.



In a way, it's a lot like the classic game of Rock, Paper, Scissors (shortened to "RPS"), since R < P < S < R, just like above. But with Fighting Yomi there's some weighted values behind each one which makes it more complicated than a simple random guessing game, because some correct guesses are more rewarding than others, or some incorrect guesses are more damaging than others.

To explain how this works in Fighters, and why it's lopsided in Fighters, I'll refer to my previous example of Alice 66B against Suika and what happens after it's blocked. Now, in this situation there's a bunch of different things each player can do, and each of them is categorized as a Yomi layer. These are the things that can occur for each player after Suika blocks an Alice 66B:



- Suika layer C is to do the least risky thing, to guard by blocking standing/croching.
- Suika layer B is to perform a risky defensive action, to jump back and airguard.
- Suika layer A is to do something completely random, like 6B or 66b.

- Alice layer C is to take the correct action, which is back off and do 2A.
- Alice layer B is to perform a risky reset, which would be to do 6B or 66B again.
- Alice layer A is to perform something totally random, like an instant air j.C or sword-doll special move.

Again, neither player can react to what the other will do after blocking Alice 66B, since the following actions will be too fast to react to, the players must assume what will come next and perform their action on anticipation. And we know the formula is A > B > C > A. Indeed, the formula holds true in this situation as well.

If both players are on Layer-A, then random garbage tends to happen such as both players CH'ing each other or whiffing each other, and other such nonsense. The only reason this wouldn't result in "neutrality" is the difference between characters.



If the Alice player moves on to Layer-B, expecting the Suika to remain on Layer-A, then Suika outright loses for mashing buttons and gets an Alice Boot to the face as seen above. On the flipside, if Suika moves on to Layer-B while Alice remains at Layer-A, then Suika will be rewarded with a successful firepunch/bomb or even a jumping kick.

If both players are on Layer-B then the situation results in a neutralization, since Suika avoids Alice's follow up attacks, but doesn't really gain much in the process either.



If the Alice player moves on to Layer-C, expecting Suika to be on Layer-B, then Alice will be rewarded with the 2A countering Suika's backwards jump by either hitting it or chipping away at Suika's spirit bar for a guard crush, as seen above. On the flipside again, if Suika moves to Layer-C expecting Alice to be on Layer-B then Suika is rewarded with a correct block, which is in Suika's favor since Alice loses control over the situation and gains nothing.

But, here's the kicker. What happens when Layer-A and Layer-C collide? Well, that's what I like to call Negative Yomi (totally arbitrary name). As in the formula before, Layer-A wins. Even though Layer-C should have been the best choice for both players, it still loses outright to the random "stupidity" on Layer-A. These are the exchanges where good players on Layer-C are often left baffled and question "Why would you do that, don't you know what could happen?" to their opponent.



To break it down, if Alice moves up to Layer-C, expecting Suika to be either on Layer-B or Layer-C, Alice will get hit by the random nonsense spewed out by Suika Layer-A like Suika 6B or 66B, since her 2A loses to Suika 6B and 66B as seen above. And on the flipside, if Suika is on Layer-C and blocks, then Alice is rewarded with a bullet lockdown into guard crush. Technically, neither player should still be on Layer-A since it's the least rewarding, and most risky, of the three layers. But alas, because it technically works, it can't be counted out entirely. It can be used to force the opponent to recalculate and readjust to the different Layers.

When you use all three options equally then you would in theory become unpredictable, and therefor unreadable. However, that also means that you're taking rather large risks that gain low rewards in the process, while the other player is not. And you would be doing this all just for the sake of throwing your opponent's game off. This is why you most often see Layer-C being the most used in tournament play, since it's the least risky and allows you to evaluate what Layer level your opponent is on.

With the above scenario, if the Alice player is on Layer-C all the time, they they are rewarded quite a bit for successful counters to Layer-B and are not harmed when the opponent chooses Layer-C as well. Now, the Alice does indeed get hit with random crap from Suika's Layer-A, but the payoff isn't nearly as good for Suika Layer-A>Layer-C as it is for Alice Layer-C>Layer-B. So even if you go through the trouble of using all three Layers, at the end of the day the payoff for Layer-C, when used consistently, ends up being better and stronger than the payoff of Layer-A. This is why many good players get frustrated when they bump into novice players relying on things somewhere between Layer-A and Layer-B. It forces them to "level down" their game to Layer-B and/or play even more aggressively than they'd like to.

Finally, you can't really classify all this as a Mixup, since a Mixup is defined by one player having absolute initiative and the other player being forced to defend against it. This also isn't a Mind-Game since again you don't have full initiative, and you're not technically forcing the opponent to do anything. This may be labeled as a Metagame since it relates to tactics and strategy, and is certainly preemptive. But for the most part, the Fighting Game community just labels it as "Yomi". And refers to bad Yomi as either "RPS" or "Mashing" because RPS is commonly seen as just random guessing.

In closing, I would like to say "anything truly predictable can be countered". However, reaction and execution time can get in the way of this. It's both the game's rules and the player abilities that determine what is and what's not rewarding/reliable per scenario.

---

- Copyright © Xenozip.