Street Fighter 4

[buhwhyen]

Pretty sure Blanka will go down to A tier. Hes pretty good in this game, but his ultra/super are really just shinshoryuken ala CVS2 (chip moves). I don't think there's a good way to land them regularly. Rainbow roll is good, and annoying for fireball characters, but I think the issue is more that Japanese players haven't gotten used to how he can move around the screen yet. I don't think blanka ball is safe? I forget, but I think I was able to punish it with Sagat on block (might have just been the guy was retarded and didn't block, though).

I wish I could find more C.Viper videos on youtube, but its so flooded with mentally retarded american players spazzing on the computer. So I really don't know how she's been played recently, though I don't understand how shes S+ tier.

Akuma...I understand why hes S+ tier. Hes a better version of Ken. He actually doesn't even need his super/ultra to do ridiculous damage, his regular bnbs do good damage are easily hit confirmable off hella stuff. His forward double kicks (rh I think?) have such stupid frame advantage and can be linked into a jab x hurricane, srk combo. Air fireball also is similar to the STHD remix version, where it does little damage itself, but gives way too much frame advantage on block (ie, free hit confirms). Thankfully, raging demon is very jumpable, similar or even easier to jump than CVS2. Also, red fireball should not be 3 hits. They more or less evened Ryu's fireball with Ken's in terms of recovery and damage (but Ryu's EX knocks down, Ken's doesn't), why is Akuma's fireball still better?

I'm surprised there's 4 characters in S+ tier. I don't think that there's a real consensus of who belongs in which tier yet. I'm pretty sure at some point in the near future, the tiers will stretch to 5 levels, because right now the distribution of characters in each tier is way too even. 8 in the top 2 tiers, 9 in the bottom 2? I don't think the game is so evenly split.

[Pikachu]

Not that the bottom 2 are even that lowly rated.

Blanka can supposedly link (?) into his Ultra from what I've read. I haven't looked it up in character specifics yet. The ball is hard to punish for some characters. They created the usual list of "How to punish a Blanka Ball" at the start of SF4, and some characters had only Ultras/Supers to deal with it. Steering the Rainbow Roll is quite hilarious, btw. Lots of free crossup damage. He turtles really well, however, and you almost have to go after him. His damn crouching move allows him to stay under fireballs even, so unless you're Sagat, Blanka can force your hand. My first game with Blanka, I purposefully ran down the clock to see how easily I could do it (and get the medal for doing so... damn medals). Probably doesn't say much since most people online are meh, but I'd be less confident doing something like this even with Bison.

Try doing a search specifically for Gamechariot videos. There have been some good Vipers there. I have some ideas on why some of her mixups work so well, but I haven't played enough with her to know if I'm right. She can apparently cancel the start of her super jump into Ultra, so that's definitely something that ought to propel her up in the tiers. I don't think she'll be S+ permanently either.

Ben did a list of Akuma combos like he often does for new games. It's some real obscene shit.

[phetus]

AAARRRGGGGHHHHH. Fuckin' bullshit. I'm stuck in this hellhole with internet for the first time in 10 days and I can't play any SF4. Bleh. I can't wait to get home. I can't even do e-mail from here. Everything's freakin' blocked by the government. Anyways, just wanted to say hi to everyone and bitch about how I can't play SF4... I still have to buy my copy when I get home. I think I get home Sunday night.... but that'll mostly be taken up by lots of food, booze and sex I think. At least I hope. Haha.... Lemme know if any of you have some free time during the week... I'd hate to go yet another week before hanging out with some friends. Lata.

[Pikachu]

Against 90% of the people online who twitch way too much, E.Honda = hilarity. I tried really hard to farm for hate mail by doing Butt Drop crossups (Short Butt Drop crosses up when done point blank). Some people who didn't know how to block the other way literally died by sumo ass. People who did know how to block would die to tick Oicho throws.

Honda Headbutt does stupid damage. There was a match I played where I hit two headbutts in the beginning of the match, and Jon's reaction was, "Uhh, where did his life go?" I did about 30% damage just on two hits. Ghey.

I tried really hard to be cheap with Honda, abusing dumb stuff... and nothing worked. However, I am proud to announce that I got my first hate mail. It's not much hate. Just an empty message with a subject line that reads, "ur defanitly (sic) gay". And you know who I used to get hate mail? VEGA. As in CLAW Vega, not Japanese Dictator Vega, which for me might have made sense.

Of all the games to get hate mail for Claw! I tried mashing 360's with Gief, retarded Scissor Kick antics, E.Honda turtling, Blanka turtling, Shoto fireball traps, Ken f+mk abuse, and beating people with DAN... and it all failed to generate an ounce of irritation! And Claw, in all his crappy glory, gets hated on? Someone explain plz kthx.

[buhwhyen]

Play against more Mexicants. Otherwise, double perfect more pplz or abuse saving attacks more (this actually annoys people a lot). Your problem is the things you would be frustated losing to is different than a scrubs. What would a scrub be frustrated by? Something he deems to be just unfair, or cheap. If you need examples, go look at gamefaqs boards? Btw, have you gotten a lot of "hey ur good" messages/friend requests? There's also people who do not understand no homo, and message you with stuff like, "ur good we should play together." I've double perfected people and they message me "I want a rematch, you got lucky." People...sigh.

Something I've been thinking for a while, but didn't want to say anything till I tested it more. This game is WAY too beginner friendly. Crossups are almost always useless against shotos and/or characters w/ srk or similar moves. Say you got knocked down you're facing right, and this is a Ken vs Ken matchup. Your opponent crosses you up w/ j.mk, you input a srk (as if you're still facing right, even though you got crossed up) and your opponent crosses over to the left before you get up. If you still input the punch on wakeup, your srk will track, and come out facing left. This was doable in previous games, but only if the crossup was just as you got up. In SF4 if you get crossed up before you get up, inputting a srk in the wrong direction still results in a srk coming out. Lame? Hell yes. This means you can 90% of the time reversal a crossup, no matter how well its done.

I'm pretty sure this is a result of more lenient buffering time for inputs. Sort of like how you can store Chun-Li's super in ST, or Honda's ochio throw/super, but not quite to that extent. If you buffer a motion, there's a certain amount of time that the motion is stored awaiting a button input to activate a move or time between button presses to recognize multiple button moves. This time in SF4, as we know from random EX moves resulting from piano'ing inputs, is a lot longer than it has been in previous games and frankly longer than it should be. On wakeup it seems like this leniency is even more lenient (hella dumb way to phrase it).

On a similar note, reversal timing on wake up is hella long. Getting reversals on wake up in this game is very easy, even during online play.

There's also a lot of talk on SRK about frame data. They were claiming that the game runs at 60 fps, but the animation is running at 30 fps. This...made no sense to me in terms of how they were claiming that it affected game play. They said that this meant that moves listed at say 3 start up frames was actually 6 frames. Lol? Thinking about it, I conclude people are mentally fucking retarded. This is basically saying that the fastest moves (not including throws/command throws) in the game come out in 1/10 second (6 frames), kara throw input windows are ~6 frames (should be ~3), frame advantage is actually more like +10 (1/6 second) off Ryu's f+mp!? Please do not listen to this mental retardation.

[Pikachu]

You know what? You're right. The stuff that'd make me irritated is stuff that would make no difference to a scrub (I fucking hate j.hk, sweep). I tried spamming single moves, but it occurred to me, after reading your post, that I'd still be doing stuff that scrubs wouldn't usually do.

1. I walk and dash.
2. I cancel a lot of shit.
3. I punish the best I can.
4. I have long attack strings.
5. I use non-heavy normals with frequency.

Shit, this is hard (no homo). I don't perfect very often at all because I walk up to block specials so I can punish and take a moderate amount of risk. It's not something I wanna go out of my way to change.

I have gotten the hey ur good messages before. Some of them have in fact been GG's that I exchanged because it was a decently competitive match.

The frame data thing has me confused a bit. I read about that the day it came out. I think it has to do with the source of the frame data, rather than how frame data actually works. Either way, it's all relative no matter how you multiply the numbers, so long as it's consistent all across the board. The charge timer argument does make sense, though, so I guess we'll see the answers unravel eventually. Ed Ma isn't a dummy, though, so while I'm not jumping on any bandwagons, I'll pay enough attention to see what the results are. In the mean time, a lot of the frame data is useful enough where I know what's fast, what has advantage, and what points I need to be careful of. If anything, I really want to see a source that lists wakeup times and properties of quick getup for each character. I'm very surprised this information isn't out yet or seems to be overlooked, seeing as how a lot of this info was prepared to be released to the players.

[sumason2]

I love it, Every time I play someone I try to throw, Focus attack, then throw... Just to see if they suck... I'll do it until they learn to jump or hit me out of it.

Then I'll do it again, but cancel the FA into a backdash/block... trying to bait the ultra, which 90% of the time works.

but I hate my d-pad... I can't punish for crap.

[Pikachu]

I actually haven't found too many people who fall for the post-throw level 3 FA for an entire round. I'll use it every now and then if they hesitate expecting a level 2 meaty.

[sumason2]

You have to get the people who play around 3-5pm our time, East coast/kids...

Yeah I know it's a little wrong, but who cares it looks really funny.

[Pikachu]

Can that be considered child abuse?

[buhwhyen]

Hyper armor only absorbs 1 hit, right? If that's the case, then Balrog's (boxer) TAP definitely has invincibility and not hyper armor. EX tigers whiff completely, and I thought you took damage but slowly recovered it over time w/ hyper armor? Or is that only w/ saving attacks?

And tiger genocide -> tiger destruction (super->ultra) is dumb (and useless). Similarly, fake kick is ball breakingly difficult to do.

And wtf I did cr.lk x3 x tiger uppercut and negative edged the super. The more I did this combo, I realized just how long the window for inputting a negative edge is. If you release the button at the same time (or close to) as the fierce punch input, 90% of the time you get a tiger uppercut canceled into a super. I really wish this would be patched, but I know its not going to be.

WTF. Hard trial mode combos are omg wtf lolol. cr.mk x low tiger shot, ex focus attack, cr.mp x tiger knee. How the fuck do you combo off a focus attack (that's level 1 and not CH). By the way, this is where you really see how useless the prima strategy guide is. It lists certain frame data, but nothing really useful. It lists the advantage/block frames on focus attacks, but nothing regarding dash cancels, dash properties, anything I'd want to know.

[Pikachu]

Super/Hyper Armor is a property of Focus Attacks, which is the one-hit absorption that you recover over time. Other moves with Armor apply with the life recovery. Armor break moves will stop them the same way.

I do c.lk, c.lp TU so I don't negative edge anything funny. It feels really unnatural to me, however. Otherwise, from what I have observed, people link the c.lk to TU instead of cancelling, kind of like what happened with CvS2.

Some of the hard trial combos are... not hard. Others make you question things. The combo you listed is the one I'm stuck on for Sagat right now. Did you do the first hard trial for Ryu btw? Jump roundhouse, low strong, low strong, sweep. Once I had a feel for the timing, I got it, but on paper, that is such a WTF combo.

[buhwhyen]

Link c.lk into TU? According to the frame data that's a 1 frame link...+6 on hit, 5 frame start up on TU. I am a disbeliever (though I have seen absurd consistency on 1 frame links before). I think they just hit confirm it like every decent Japanese player can. Short short, short jab...either way its enough time to hit confirm if you're looking for it. My problem with hit confirming is I buffer the motion no matter what. Why is this a problem (almost everyone does it this way)? I seem to get a lot of unwanted moves because I either: negative edge b/c I hold down the last button before the move I want (situationally, not always), or the motion gets buffered and it gives me funky shit (read: special move or super) when I try to do for instance walk forward tiger shot or walk forward insert normal move because I don't pause long enough before putting in the next input.

By the way. If you really want to see hilarious buffering time bullshit. Do a dash (off of saving attack or just a plain dash) and buffer a super/ultra really early. I need to test it more, but when I was doing this the other day I inputted the qcfx2 way too early. Normally I'd just do another qcfx2, but I choose to not and just push the buttons when the dash ended and got my ultra to came out. It seemed like there was almost a full second or so between the motion and button press (thx slow Sagat dash). It felt SO awkward, but totally absurd how forgiving the buffer time was.

dash, qcfx2, pause (long enough that you could input another qcfx2), 3 punches = hilarity. I think this is why scrubs can so consistently pull off wake up srk w/ Ken. Mash motion (as long as they get 2 qcf somewhere in there it'll recognize it as a srk, even if they mash out a 1080) + mash many buttons (something will probably come out, since they don't have full meter = regular/ex srk, if they only mash fierce, fierce srk comes out, otherwise probably ex meter depending).

[Pikachu]

I will try that tomorrow. That makes a lot of sense, actually, because it's a good explanation for Ultras randomly coming out in-between my poke strings when I lag. If they just spaz enough, there will be enough inputs to qualify for a "motion" and hitting three punches is definitely not too hard to envision.

In some games where I think the person is a real spaz, I've actually stopped poke strings halfway to allow for these moves to come out so I can punish, but the results are random. Guess we still have to be careful of scrubs because they can bite you with Ultras out of nowhere; if this is true, then the game truly is scrub friendly.

[Pikachu]

Just as a warning, this is a post on a lot of random thoughts to draw a conclusion that is basically... DUH. At the same time, I'm just throwing it out there as a definitive argument/reminder on why sweeping/throwing is such a crap punish.

So I was thinking about situational minutia - stuff that comes up often enough to think about, but not often enough to have a breadth of knowledge on - for SF4 after trying out random shit while playing Alvin the other night.

It got me thinking about the post-Hurricane Kick scenario after Ken lands it on hit and the 50-50 it presents, which my stream of consciousness got me wondering why Ken scrubs, for all their spazziness, still win. If we step back and think about it, Ken scrubs do win more than they lose, even if all of us have a winning record against them. That got me thinking about what kind of conditioning it might take to stop random, seemingly insane Shoryukens.

We should know that for a beginning player, it'll take a lot more than one game to pound insight into why a strategy loses, even if it's seemingly obvious to us. I am starting to believe that Ken players aren't being punished enough, or often enough, to make them change behavior. It was on this basis, that I hit a bit of a hypothesis that I want to see if it is a viable line of thought.

Let's start with a control situation: the match is Ken1 vs. Ken2.

Ken2 jumps in, but Ken1 pre-empts the jump with a j.Fierce. Ken2 will reel in the air for a bit and land on his feet when he hits the ground a bit after Ken1 lands. We'll assume Ken2 is also San Jose State Alumni, so he is going to Fierce Shoryuken when he lands.

Situation 1: Ken1 can try to put a hitbox (let's say a low roundhouse sweep) where Ken2 lands in order to start some offensive momentum, but the Shoryuken from Ken2 will hit Ken1, and cause 140 damage.

Situation 2: Ken1 blocks the Shoryuken, takes a bit of chip, and punishes with a a sweep or throw, the two most common methods of punishment.

There are more potential situations, but we're talking about pure Shoryu-spam, so I'll limit the scope to this 50-50 "game" (and I need to keep it simple because I'm a psych major, not an all-out game theorist).

I think we can agree from experience that if you know a person graduated from SJSU with a BS in Shoryukens, it's a good idea to block in this situation more often than not. But for a lot of people who are a bit more random in behavior, I think it's okay to assume a coin-flip.

With the assumption of randomness, we can look at the damage expectation of Ken2's (the Spammer's) risk, and the numbers make a good case for him. On success, he gets 140 damage, 240 for EX when he gets the meter. The person doing a sweep or throw to punish, however, is rewarded with 120 on the sweep, 130 IF you do a back throw, which is the less common throw. While 140 is only 20 more than 120, the Shoryuken spammer has a 116.67% advantage on his damage expectation each time in this situation, which puts his risk vs. reward ratio at an advantage.

Now if we add the human element to Ken1, he can still outplay his opponent and come out on top, even with a sub-optimal punishment option by reducing his risk, especially if we still consider that Ken2 will ALWAYS Shoryuken, being the paragon of a scrub's wet dream.

The pre-emptive Shoryuken is kind of like a random Shoryuken, in that there is some guessing involved. Situations where a Shoryuken will be thrown out to stuff an attack happens quite a bit even in competitive play. Assuming mind games on equal skill level, we can return to the assumption of 50/50 with a third situation thrown in, and that's if both players choose to block momentarily, which gets neither player any damage, so I'll just ignore it in terms of calculating damage expectation.

Against a sane player, it is sometimes alright to stick a move out in the ambiguous Shoryuken situation because you know they won't spam it all the time (unless it's a SJSU player who is otherwise good at the game, but cannot help the qualities of his alma mater). This is a good choice to have, because a player wants to have the potential for offense open to him, whereas blocking might simply result in giving up a momentum shift...

...unless the player about to go on the offense will only punish with a sweep or a throw. Putting a limb out against a Shoto in these situations is always a risk, and here, the risk is 140 damage. The person who wants to Reversal Shoryuken only has a risk of 120. The person sticking out a limb is rewarded with a bit of block pressure. The player uppercutting gets a knockdown. That's quite a difference, especially if this situation is played out a few hundred thousand times over the many games people play. If we apply this into the world of scrubs, the Ken spazzers are making the correct decision by uppercutting every time because they literally are not being punished enough (i.e. not having their risk raised enough due to the inferior damage they're punished with). Imagine how much more of a damage swing there is when they spam supers/ultras!

Even then, if we are to apply this idea to solid play, we should see that we need to earn the offensive pressure. Again, we'll assume equal skill causing a zero sum mind game:

Critical Shoryuken Point #1: Mitch (a white sounding name), who is Ken, plays Jose (a Mexican sounding name) who is also Ken. Jose spazzes a Shoryuken (an incredible feat for a Mexican, who are known for simpler uppercuts), which Mitch blocks.

Mitch punishes back with target combo into Shoryuken, dealing about 290 damage (maybe closer to 270 with some damage scaling).

Critical Shoryuken Point #2: Mitch tries for a meaty attack on Jose's wakeup, but Jose lives up to his saintly standards and spazzes another glorious uppercut, hitting Mitch for 140 damage.

Now in the guessing game, the guesses are even, but Mitch is ahead, and Mitch is not "wrong" like Ken #1 was before for sticking out a limb, because he has increased the risk for Jose to be throwing his uppercuts out. The difference here is all in the punishment, because it is the potential for punishment that sets up the risk/reward standards for the match at all. What I find interesting about this line of thought is how much it can affect gameplay on a large scale. In individual games, the results can vary, but over a high number of games, the damage expectation will swing in favor to the otherwise high risk options (because the punishment for being incorrect is not enough to actually make it high risk). There are also many other similar scenarios, like throwing random supers, spamming piledrivers (a very low risk option for Gief), Buffalo Headbutts, Flame Kicks, Cannon Spikes... you name it. That's a lot of risk/reward scenarios being thrown around.

A thing that has struck me is how this is so relevant to American and Japanese playstyles (at least in general). Japanese players have commented time and time again that they get caught offguard at how offensively driven so many American players are (against our good players). On the other hand, if we are to watch Japanese videos of almost any fighter, they stay so ridiculously safe. Why do Americans insist, after all these years, on spazzing uppercuts and supers when we know better? Even some fairly good players do it in addition to all the bad ones. My argument is that when we play on American soil, I don't think we (as a country) may be making it a wrong enough decision.

Of course, on these statements, it's hard for me to say how concrete it comes into play, or how precise the variance is, given the complexity of fighters in general. But I think there's some general sense in it. I purposely checked out scrub videos on YouTube that light up with flashy, spammy Ken moves, and sure enough, the ones that are blocked at all are always thrown or swept. It's not like anyone should needs to calculate damage expectation to know that you really ought to smack the opponent as hard as possible if you get the opportunity, but the idea makes for an interesting longitudinal study in conditioning.

And maybe some concrete info has to be thrown in people's faces to convince them to remember to punish harder.