Fighting Game Tool-Assisted Speedruns

[Maj]

Changing one command in a sequence takes seconds on a program pad, too. Emulator tools are more powerful than a programmable pad, but in most cases the two are comparable - only separated by a matter of degree, not in entirely different categories.

I see what you're saying about that COTA Storm combo, but what if you get to the last hit and something doesn't work out, forcing you to restart the combo with Storm on the right side? Or with slightly more meter? Or with a less/more damaging dizzy combo? Or against a different opponent? Or on a different stage? Again, i think the lack of competition is making things a lot easier for you. If you choose a game that people care about, then you create a much greater challenge for yourself. The problem is that you don't have an objective, because you don't have a precedent to overcome, so you don't feel the demand for innovation, so there's no pressure.

CvS2 has at least half a dozen uniquely significant combo dummies. You can't just pick any two characters out of the blue, start a combo, and end up with something presentable. You need planning, creativity, foresight. No tool is going to do that for you. Trial and error will take you weeks. Believe it or not, the same is true of MvC2. I may not know all of the limits in that game, but they are definately there. Even with the best emulator tools, it would be very difficult for anyone to make a Magneto combo that Magnetro couldn't improve after watching it once.

I like being a fan of sports players and video game players who have specific styles and are very talented and experienced. It's fun rooting for your favorite team/player and being entertained and awed by their display of skill. It's nice to have a role model/hero. But, I don't think I'd be much of a fan of Robot#2501, but I can appreciate the thought that went into it's programming.

That's a good point. I definately enjoy looking forward to videos by my favorite authors. Though, if you think about it, a lot of the people we admire use programmable controllers. Sai-Rec, Tosaka, zerokoubou, ozy, joo, kysg and on and on. TZW's methods may not have been nearly as sophisticated as the tools available today, but he definately used rapidfire and who knows what else? The reason i'm a fan of Sai-Rec is because they're genuinely innovative. The video clips they released in the early days of CvS2 were absolutely groundbreaking, and so was the first full-length video they sold. The second one went in kind of a boring direction, but that incomplete 21:16 video is all sorts of creative. Seriously, there are at least three or four 5-minute CvS2 combo videos made by Japanese players using programmable controllers, containing only one or two minor innovations. Then there are the "REQUiEM" videos by T-7 which make huge improvements over the Sai-Rec stuff. Clearly, there's an issue of skill at work here, because lots of people with programmable controllers never come up with anything groundbreaking.

Problem that I see is that people have generally already made up their minds. The way we handle things counts for something, but it's probably not going to make the majority of people suddenly change their views on the subject.

The fighting game community is sort of an odd beast. It's not at all like the other communities I've been a part of.

Yeah, the fighting game community is weird sometimes. But what can i say? I really dig it. So i'd like to think that i'm having a positive impact by contributing articles, videos, and stuff. A lot of people have strong opinions, but i think a lot of people are open-minded and flexible too.

The most dangerous thing is becoming elitist. That's what we gotta watch out for. Cuz from the very beginning, my primary target audience for my combo videos has always been a very small group of maybe five to ten people. They're the ones that i cared most about impressing. Sometimes i don't even know why i bother hosting a website, instead of just sending those videos out to like ten people. But as long as i think it's somehow helping the fighting game community as a whole, i'll gladly take that extra step.

The fighting game community needs combo videos to keep things fresh and exciting. Making tool-assisted combos is kinda like taking a big step forward, and it might be very difficult to undo that step, but i think it's time we made the jump. Besides, we've been a step behind Japan for years and years.

No matter what, match play will remain the most important part of this community. If you ask 100 Marvel players for their favorite Magneto combo, i bet like half of them will point to SooMighty's snapback combo against Justin Wong's Sentinel. And that's perfectly valid. And that will always be there. We don't have to be afraid of breaking anything fragile, or handling everything with kid gloves all the time. Nothing will undermine the value of execution under pressure.

[Magnetro]

Tutorial videos help sometimes. I think there are things that no one understands about marvel and wonder why they happen that would like to see some of those things explained in a video.....

There are like 2 people in SRK right now that are trying to find data for small tiny stuff to use in a match not for combos, haha. So maybe a video explaining common stuff in MvC2/CvS2/lesser games will be helpful and get people's attentions.

[Maj]

Actually i think the CvS2 and 3S tournament scenes are way more technical than the MvC2 players. The CvS2 players are always talking about frame data, which characters get up faster, which characters have the best throw ranges, what can be punished after Just Defending, and so on. There are definately some technical MvC2 players, but i think that on a scale of ghetto to Virtua Fighter, MvC2 is about a 4. You know, it depends on the game. For some games, players can't get enough random obscure technical knowledge.

[Magnetro]

Actually i think the CvS2 and 3S tournament scenes are way more technical than the MvC2 players. The CvS2 players are always talking about frame data, which characters get up faster, which characters have the best throw ranges, what can be punished after Just Defending, and so on. There are definitely some technical MvC2 players, but i think that on a scale of ghetto to Virtual Fighter, MvC2 is about a 4. You know, it depends on the game. For some games, players can't get enough random obscure technical knowledge.

Yeah, MvC2 is a lot more fun than other games.

As for the discussion: It's really hard to make certain combos work with a program pad. It's never as easy as I'd hope. Certain combos just don't work 100% of the time and if they're super long (999 hit combo/Secret sent combo/certain resets) there are a lot of times where it's hard enough to figure out the general numbers to programming a certain sequence but then they don't work 100% of the time and you can't do anything about it or else they won't work! That can be annoying especially since you can't do it by hand. I dunno, it's hard to explain unless you've tried programming something in a ppad. Heh, ironically that magneto video I made like two years ago wasn't that hard at all, =/.

[jchensor]

My personal reasoning for appreciating manual runs more is: watching manual runs you're seeing the application of skill, thought, and effort. With TA' you're only seeing the application of thought. I personally like the idea of using TA' as a template for manual runs -- to show what is possible and what to strive for.

I actually disagree with this. I think Tool Assisted stuff requires more thought. Manual things are all almost exclusively skill, muscle memory, memorization, and emotion. Basically, manual means you are given a set of things you can do, say, A through P. You avoid Q through Z because they really are too hard to do for a human. So you stick with what you know and what you are good at to minimize errors.

So what happens with Tool Assisted stuff (and this is in reference to Combo Videos in particular, not speed runs), you now have A through Z to work with. But now, NOW here comes the hard part. If you have only decent piano skills, there's not much you can play. But if you have maximized piano skills, you can play whatever you want. But you wouldn't want to do so, right? Only ceertain things will truly sound beauitful. So when you have A through Z to work with in a Combo Video, your goals change. You aren't looking to push the limits of a Combo anymore necessarily (in certain games). You are looking to accomplish certain goals. Or have things happen that are just too cool for words.

kysg is a perfect example of this. I love love love his videos. And it has nothing to do with inhuman links and impossible kara cancels. It's because he fucks around with Combos so much and tries to do things that are like mini-goals for the sole purpose of being stylish. In the Ryu video, he tries to get as many Towards + Strong overheads in one combo. He also lands the one Combo where Sean bounces on, like the Super Fireball, then the EX Fireball, then he Joudan Shoguto Geri (sp?) kicks him. That combo was awesome. In the Alex video, he does 100% damage combos with each super. He does one combo where his only goal is to land as many Stand Roundhouses as possible. In the Necro video, he'll end a Combo with a Juggling Sweep only because it's cool to see the Sweep Juggle. If he went another route, he probably could have landed 2 or 3 more hits. But instead, he chose to do something stylish.

And that's where it all starts to work out. TA Combo Videos aren't about making the best Combo you can make. It's about being stylish, about turning Combos into art. Back to my earlier analogy, anyone with infinite piano skills can try to make a piano sonata by playing a TON of notes and playing them all as fast as humanly possible. Yeah, it's skillful, but it will probably sound like crap. But, because they know what they want to try and acheive, the songs they play will be more beautiful than that, with far less notes and they might even be slow. ^_^

So that's why I think it actually takes more thought to make Tool Assisted Combo Videos. Your goals have changed and it's up to you to execute your creativity. This may sound like an oxymoron, but it's the limitation of limitless freedom that you have to overcome now. It's easy to make a combo like the Storm Combo you did, sure, but I do have to agree with most that say that COTA is broken beyond all reason (a Combo Video would be more impressive if they showed all the things that DIDN'T work. ^_^ I think there are less of those than stuff that does work). But see, once you have that power, it now becomes a question of what do you choose to do with it? If you had that power and you decided to make a Storm Combo where every time she built up enough Meter, she ONLY used the wind blowing and still managed to keep the combo going somehow going back and forth between corners, that would be awesome. Then people will see it and go, "Oh shit..."

It does depend on the game, largely, as well. Games like MvC2 games, where the limits aren't as strict as CvS2, style is huge. But in games like CvS2, where you really don't have that many options, then maybe maximizing hits could be a valid goal. You really have to judge the game and figure out how you wanna make the Combo Video. But with unlimited power, you need to make those goals for yourself in your mind.

- James

[Xenozip.]

-snip-
- James

It's entirely possible. And yeah, I don't really like repetition. Either performing or watching. What I really like to see is wacky and/or variety. If a combo contains everything that's possible just once, rather than the same thing 5 times in a loop, then it becomes more entertaining to me.

I liked the Ode to the 2 Hit so much because it was so wacky and you never really showed the same thing twice. Of course, I'm guilty of repetition myself, but that doesn't mean I liked it.

But in response to both you and Maj regarding the clip itself: The point of it was to just show how retardedly easy it was to spew out something like that in less than a day without having played the actual game for any serious amount of time. Take that for what it's worth.

Of course it's not a super amazing video, I barely put any thought or effort into it at all. I wasn't trying to make the best combo out there or compete with anyone. I'm not really interested in competition on that level. I don't disagree with what either of you are saying, but all I was saying is it takes away the player and leaves the brain. Only point of making the video was to show what I meant on a basic level. I hate to repeat myself again, but just for the sake of emphasis, it really did take me less than a day to make those clips.

[Goryus]

I find the divergence of opinions in this thread fascinating.

Personally, I do not think the purpose of a combo video is to show off a player's skill. The purpose of a combo video (or a TAS, for that matter), is to show off all of the interesting things that can be done when the game engine is pushed beyond what is normally practical.

Whether hand-performed or not, the combos in any decent combo-video you choose to name are probably not practical for match play. They are the results of hard work, knowledge of the game engine, and ingenuity, not skill at playing the game. As a result, a combo video created with tool-assistance is not really analagous to a team of robots playing basketball. It's more like watching a hyper-realistic, humanoid robot doing crazy, unexpected tricks with the ball, made all the more insane by the fact that the robot is working under the physical limitations of a human body - just making use of far greater precision than a human is normally able to muster.

Because I interpret the purpose of a combo video to be showing off what is possible in the game, and not a demonstration of personal skill, I am in favor of pretty much anything that improves the result. That includes emulators, programmable controllers, and the good-old fashioned "try it a billion times and only use the one shot that actually worked in the video." Even if doing so may be misleading in terms of how "skillful" I was in order to pull it off.

[Maj]

As a result, a combo video created with tool-assistance is not really analagous to a team of robots playing basketball. It's more like watching a hyper-realistic, humanoid robot doing crazy, unexpected tricks with the ball, made all the more insane by the fact that the robot is working under the physical limitations of a human body - just making use of far greater precision than a human is normally able to muster.

That's the subtle fine line, isn't it? There are some things that the video maker has to decide themselves. For example, there are some tool-assisted speedruns that incorporate pressing two opposite directions simultaneously to create movement glitches. You couldn't do that on a Genesis controller, but you might be able to open up the Genesis controller and press the triggers for L and R simultaneously. So where does the video maker draw the line? There are definately some murky areas. Personally i think pressing two opposite directions at once is crossing the line, but certain issues are way more subjective than that.

In principle i totally agree with you, that even manually performed combo videos are deceptive. Even when people talk about "what's humanly possible" versus "what requires tool assistance" ... that line isn't as solid or as important as it sounds. Most of the time when i'm doing complex combos using elaborate setups, i never feel like i "mastered" the combo. I simply get to a point where i have a general grasp of the timing, and then i repeat the same thing 200 times until i get lucky once while recording. Is that really skill? What does that really prove? You can call it persistence but it's certainly not finesse.

[jchensor]

But in response to both you and Maj regarding the clip itself: The point of it was to just show how retardedly easy it was to spew out something like that in less than a day without having played the actual game for any serious amount of time. Take that for what it's worth.

Of course it's not a super amazing video, I barely put any thought or effort into it at all. I wasn't trying to make the best combo out there or compete with anyone. I'm not really interested in competition on that level. I don't disagree with what either of you are saying, but all I was saying is it takes away the player and leaves the brain. Only point of making the video was to show what I meant on a basic level. I hate to repeat myself again, but just for the sake of emphasis, it really did take me less than a day to make those clips.

yeah, I can see what you're saying. I never did mean to imply that you were going for some super bad-ass combo in COTA. I guess my point is that the engine is so free, it's a good game to be able to make even a decent combo like that with minimal effort. But plug in Alpha 2 instead, and I wonder how tough it would be to make something even remotely decent. Basically, those games are at a point where only the exceptional will do. So unless you are doing Zerokoubou quality stuff in Alpha 2, it just doesn't seem like it's worth trying anything in that game anymore.

But you are absolutely right in one thing: the "human" factor is eliminated. It does become a 95% brain type of thing. So it does become a very different thing. Then it's just a matter if it works for you or not. I guess it's more of a thing where I've done enough Combos that make me angry that I can't be 100% positive whether the combo just isn't actually possible or if I just suck. ^_^ So it's nice to be able to say definitively whether or not it'll work. Helps me sleep at night better. Hahahaha...

- James

[Toxy]

This is a bit off topic but i don't want to start a new thread just to say one comment so i'll just post it here for now, as James was basically saying that Alpha 2 combos are very hard to think of, I think its probably the Street Fighter game with the least amazing combo videos for it, i mean most A2 videos are just jump in > few ground link > super etc.. or just Custom Combos, the only really fancy type A2 videos i have seen are the zerok' videos, so what im wondering is A2 the game with the least good combos in it, or do you think A2 might have a lot of amazing combos in it but people have just not found them? I'm thinking option #1..., but hopefully there are some more crazy combos in A2, I'd just love to see them since I've seen so little from A2 so far.

[Maj]

Personally i've never felt any deficiency in the SFA2 combo engine. It's a solid Street Fighter game with a diverse cast of characters. It doesn't have the chain combos of SFA1, but those were boring anyway. On the other hand, it does have very interesting CC setups, cuz you can throw a fireball and run after it very easily.

The problem with A2 is that it came out at a weird time. Back then nobody had capture equipment. When A3 came out, that's when everyone started getting capture cards. Even though America liked A2, everyone converted over to A3 eventually. Keep in mind that SFA3 is quite possibly the most intruinging puzzle of a combo engine that Capcom has ever built. So all the game engine fanatics were totally absorbed by trying to figure it out. On the other side of the ocean, Japan never liked A2 so they never pursued it either.

As far as the videos go, i'm still very impressed by NKI's SFA2 videos. Then there's a couple of SFA2 videos by zerokoubou which are absolutely groundbreaking. And of course there was that oldschool Skill Smith SFA2 video, which has some amazing stuff in it but it's gonna be hard to find these days. They used infinite meter throughout most of that video so there are a lot of combos with multiple CC's used in stylish ways. In A2, infinite meter was like steadily recovering meter instead of always staying maxed, so they still had to plan everything out wisely.

Do i think that zerokoubou's videos can be topped? Yeah, definately. Do i think that it's going to happen anytime soon? No, probably not. But watching those vids, i don't know what more you can ask for from a combo engine.

[Toxy]

The skillsmith vid is up at http://www.zachd.com/mvc2 im pretty sure, was pretty good but im not really into combos with infinite meter.

[Maj]

Neither am i, but at least they managed to show a lot of interesting things. Maybe they felt that the meter limitations made it too easy to do crazy combos with programmable controllers so they wanted to make things a little bit more difficult for themselves. Anyway there are some videos with infinite meter that are absolutely terrible and there are some that are pretty interesting. In fact i think Meikyo has a fairly good track record of using infinite meter wisely without overusing it.

[Xenozip.]

Combo videos are innately similar to tool assisted runs anyway.

If you were to do a Metroid or Super Mario run and chopped it into several segments, then recorded each segment several times till you "got it right", then patched them together, that's tool assisted (video magic).

Combo videos are often done this way, record the combo over and over till you "get it right", then patch it together to make a movie.

I suck at basketball and never play it, but I could probably record myself getting 100 swishes in a row with video editing. Likewise, I suck at Guilty Gear and never play it but I could probably record myself doing extremely advanced FRC juggle combos/loops for optimal damage/hits.

So I suppose frame manipulation and save states really don't make that much of a difference in regards to combos. I suppose it's the same thing as shooting hoops by yourself.

And this is why I don't like it when people approach me expecting me to be a super good player. I'm average in lots of games, certain games above average, plenty of games well below average. My videos give off a false sense of skill. The videos I make depict knowledge, not skill.

I wonder if Action/Hero movie stars have to put up with that kind of thing. They probably do, they are probably accused of being shitty fighters "in real life" all the time, or the opposite occurs and people expect them to be actually super good fighters when really it's all movie magic and they aren't good at all.

:|

I'm tempted to start putting a disclaimer in my videos: "My execution is not this good. And I suck at this game. But, I know everything there is to know about this game, so please let me show you".

[Magnetro]

Combo videos are innately similar to tool assisted runs anyway.

If you were to do a Metroid or Super Mario run and chopped it into several segments, then recorded each segment several times till you "got it right", then patched them together, that's tool assisted (video magic).

Combo videos are often done this way, record the combo over and over till you "get it right", then patch it together to make a movie.

I suck at basketball and never play it, but I could probably record myself getting 100 swishes in a row with video editing. Likewise, I suck at Guilty Gear and never play it but I could probably record myself doing extremely advanced FRC juggle combos/loops for optimal damage/hits.

So I suppose frame manipulation and save states really don't make that much of a difference in regards to combos. I suppose it's the same thing as shooting hoops by yourself.

And this is why I don't like it when people approach me expecting me to be a super good player. I'm average in lots of games, certain games above average, plenty of games well below average. My videos give off a false sense of skill. The videos I make depict knowledge, not skill.

I wonder if Action/Hero movie stars have to put up with that kind of thing. They probably do, they are probably accused of being shitty fighters "in real life" all the time, or the opposite occurs and people expect them to be actually super good fighters when really it's all movie magic and they aren't good at all.



I'm tempted to start putting a disclaimer in my videos: "My execution is not this good. And I suck at this game. But, I know everything there is to know about this game, so please let me show you".

I just tend to ignore the people that tell me that I am a really good player or something. You shouldn't worry about that, IMO.

[Goryus]

I think the funniest ones are the guys who say, "Oh yeah, I do that too," or "You play like me, but I do something even better."

[Maj]

Combo videos are innately similar to tool assisted runs anyway.

Very interesting points you brought up. Personally i've never taken much pride in doing stuff by hand. I just looked at it as a prerequisite for getting people's attention to avoid that whole programmable controller debate. Back then we had no idea how to obtain them so it never really came up. Ironically, all the tool-assisted combos back then were made using emulators and they were all week-one SFA3 combos - the exact definition of wasted potential.

But yeah, most of my combos were simply a matter of coming up with an interesting idea, then banging my head against the wall for 8 hours until i got lucky once and got it on tape. To me, the execution side of combo videos has always been somewhat diminished. I mean, i can sympathize with jchensor when he talks about mastering a combo or finally getting a difficult combo to click. But still, if i want to see crazy execution, i go looking for epic tournament finals, not combo videos.

So yeah, great post dude. I agree with everything you said.

[ShinjiGohan]

But how would you get that execution if you don't practice? Specificly combos, which would then be aided by combo videos as you have some incentive to practice just that.

[Maj]

The problem is that almost none of the stuff i put in combo videos is master-able. It's stuff that i had to practice a lot in order to get "pretty good" at, then repeat 100 times until i got lucky once.

If you string enough one-frame transitions together, the combo stops being about execution and starts being about luck. The point that Xenozip was making is that combo videos only show the viewier the lucky instances, which is basically the same thing as manipulating luck in tool-assisted speedruns.

Now, you can say that some combo video makers do master their combos so their vids truly are execution demonstrations. However, that's personally never been the case with me. Whenever i find a cool combo that i can actually pull off consistently, i complicate it until it's beyond my execution limits before i consider putting it into a video.

The bottom line is that if someone watched my old Guile videos and complimented my execution skills, i would deny it. I would tell them that each of those 5-second clips in fact took me 8 hours to do. That my execution is only slightly above average when compared to good fighting game players. That anyone could pull off the same things if they were persistent enough. That the only truly commendable thing about those videos is the ideas and concepts that i came up with.

I'm just sharing my personal perspective. Some people really do have amazing execution and they deserve props for that. I'm not enrolled in that club though. I hope people think of me as creative, but it's never been my goal to make people think of me as having great execution.

[Xenozip.]

But how would you get that execution if you don't practice? Specificly combos, which would then be aided by combo videos as you have some incentive to practice just that.

I strongly disagree. I've proven this to myself countless times so far.

Most recently with the game Monster and IaMP. If you look at my match videos on youtube for Monster, and then look at the combo video for Monster that I collaborated on you will see two entirely different things.

I may be able to execute several 100% combos that require 50+ hits for several different characters, in training mode for the video. But in a real match with my main character I don't even go beyond very basic b&b stuff. Why? Because I only needed to record the combo done correctly one time, just once.

However, it didn't take a damn bit of practice for me to pull off those 100% combos either. I just went into training mode, tried it a few times till I got it right, then moved on to the next one. And now that I've done that I still never even use any of it in a real match. Because I didn't practice it, I just formulated a theoretical combo in my head, tried it out, got it recorded once, and moved on.

My point was my execution in basket ball is not good at all, not even in the slightest, a little girl could probably even kick my ass in horse. But give me a camera and I can record nothing but swishes and then make a video with 100 swishes in a row. Now I go back and try to beat that little girl in horse and she still kicks my ass. This is because practicing in training mode by yourself and using video editing to take short cuts is entirely different from practicing against a real little girl.

This applies to all my videos. And it will apply to my next video (MB), and the one after (3S). My videos depict knowledge, not execution or skill.

Don't get me wrong, I may be a good player in Monster and a decent player in IaMP and decent in MB, etc. But my execution is nowhere near combovideo level, the stuff I do for combo videos doesn't transfer back to real play, and visa versa.

[ShinjiGohan]

Guess I'm a tad bit different then. As I try to use most of my combos in actual matches.

Similarly my brother after seeing a combo he liked, he'd spend several weeks getting it down. Like when Bas released his Akuma combo video of the OTG V-ism combo. Same with other combos that he sees, from Makoto, Iron Man...

[Xenozip.]

I spend a lot more time coming up with strategy and tactics. I really want to learn how to play the game before I ever bother with combos.

Like, if I can't hit you consistently in the first place then what good are combos?

Likewise, if all I know are combos but I have no idea how to land them and I keep getting my ass rushed the hell down, what good are my combos?

I do spend a lot of time in training mode experimenting with combos in specific games, but none of that stuff ever really transfers over because I much rather focus on improving my game before I incorporate advanced combos into it. When I play real matches with people, especially people better than me, combos are an afterthought.

Some games I've been playing for years now and I've gotten my strategies down where I know what to expect, I know what counters what, I know all the possibilities and permutations for every situation, what people are likely to do, what people should do, what people don't often do, risk/reward ratios, I know how to anti-air/ poke/ bait/ zone/ feint/ turtle/ runaway/ beat/ oki/ pressure/ rushdown/ trap/ option-select/ read/ mixup/ etc etc etc. But I still don't bother with advanced combos until I'm thoroughly comfortable with what I'm doing.

Unless it's something that sticks out like a sore thumb as something that leads to implied damage. Like in Monster there's a character who can drain the opponents life with something like poison, so there was actually a reason to incorporate advanced setups for it because the implied damage was significant enough to merit a great deal of focus on it.

[Maj]

It's been very quiet on the fighting game TAS front. They posted another update to the MK video, but that's about it.

SNES Mortal Kombat II (USA v1.0) in 22:43.68 by M.S.M. (aka Samhain-Grim, aka VANDAL) (starring Shang Tsung)

NES Mike Tyson's Punch-Out!! (USA PRG1) in 17:50.0 by Andrés Delikat (aka. adelikat)

SNES Super Punch-Out!! (USA) in 15:59.88 by Andrés Delikat (aka. adelikat)

Not sure if the Punch Out games qualify as true fighters, but they could work if you ran some kind of a versus mode through head-to-head cabinets. Split screen would probably work too. Hopefully someone will submit another SF game speedrun soon so we can have something to talk about.

[Maj]

Not a fighting game, but rather a mediocre side-scroller with an ominous 90's ambiance that i played (but never beat) when i was a kid. This speedrun is surprisingly entertaining because one of the alternate weapons fires super-slow floating ammunition that can be controlled like Oro's EX fireball. More importantly, the player simply walks through like 90% of the bullets fired at him.

Genesis Robocop Versus The Terminator (USA) in 14:06.92 by Cardboard

Robocop's back-to-back deaths at 13:37 and (especially) 13:43 are classic!

[Maj]

N64 Turok - Dinosaur Hunter (USA v1.2) in 36:37.42 by mmbossman

Don't even try to tell me Turok isn't a fighting game, because it is prominently featured in countless classic NKI productions.

Assuming you stop counting at one.

[Xenozip.]

This isn't a "speedrun" per se, but it is tool assisted (and an actual fighting game) Samurai Shodown 4 - Bust Basara.

[Maj]

That was pretty sweet. Did you make it?

Not that i know anything about Samurai Shodown (or Basara especially) but i liked watching him toy with the AI. It's too bad SNK loves flickering lifebars so much cuz blurtube murders them. Any chance of seeing something like this for a Capcom game?

[Xenozip.]

I think this could have only been done with Basara, due to his shadow feint moves.

I can't think of any game in any other fighter at all where this sort of thing would be even remotely entertaining. I can't think of anyone in any other game with the right kind of "tools". But I'll go stare at the character select screens thread for a while.

[Maj]

Hm, how about Ryu? He can do anything! Ryu is strong.

Seriously if any game is begging for this kinda treatment, it's Alpha 3. The non-stop hijinks will be non-stop. I nominate V-Cody, or maybe Rolento, or quite possibly Karin, but definitely Ryu.

[Xenozip.]

With Ryu, what would you do?

There's no illusion moves to make you see something that isn't there, and the AI is very linear. In SFA3 he has a fireball feint, and that's about it.

The point of the Basara video was two things. First, to do things that are absolutely ridiculous and random in very high risk situations, but succeed in them where they might normally have a high chance of failure, but succeed every time. Second, to trick the viewer into thinking you did one thing, when in reality you did another -- such as jumping toward a downed opponent, landing, then using a shadow feint to make it look like you jumped away, a normal reaction to which is to believe the player jumped back, but in reality Basara was right in front of the opponent and threw right as the shadow move ended. The only other character that has anything like it that I can think of is Melty Blood's Ren or KoF's Chizuru.

I guarantee that Basara is the only one that would be interesting to do this with in the whole SamSho4 roster, the only other remotely close runner up being Hanzo Hatori. Ridiculous teleport shenanigans aren't nearly ridiculous enough though. Nor are just hitting, throwing, and comboing the opponent.

For SFA3 Ryu I don't see how/what you would do. Hitting limbs with overheads and hop kicks, throwing limbs, karacancels for no reason, airthrowing pixels off the ground. I just don't see it being the same.