OG's halp: When did the current usage of how the last startup=first active frame of a move start?

I was discussing this today with some friends after reading Novril Tataki’s arguments with someone over on TYM about frame data and it got me interested in something; the origin of differing frame data terminology, specifically the startup frames.

Here’s the issue explained:

*Ryu’s LP Shoryu in SF4 hits on the 3rd frame of the entire move, that much is fact. A 3-frame reversal.

Does it have 3 frame startup (what we usually hear, especially in regards to safe jumps) or 2 frames of startup and active on the 3rd to be more accurate?*

To be honest this nitpicking part doesn’t really matter because we all know when it hits, but when did it become commonplace to lump the last startup frame in with the first active frame?

James Chen discussed at around 8.30 in Ultrachen’s First Attack video Frame of Mind about this convention and a short (and actually good) discussion happens in the recent comments about it being ‘easier’ to understand giving that the matching numbers of frame (dis)advantage and startup e.g. -4 on block and a 4 frame startup move means that the move will hit.

I looked up some Chun-Li 3S frame data and animations from Baston because I remember hearing Chun-Li had a 2-frame startup crouch jab and I thought that was ridiculous, and while the frame data says “2 frame startup”, using the hitbox/animation tool on there I can see it becomes active on the 3rd frame, like a lot of jabs from SF4 rather than the hitting-on-the-2nd-frame jab that I expected. (Am I wrong on that for SF4? Feel free to correct me). Also in that website it states “Startup corresponds to the number of frames BEFORE an event (hit/block) is registered by the game. Therefore, a 2 frames startup move will actually hit on the 3rd frame.

TL; DR Title was enough:
So basically I’m wondering when that terminology originated? Is it a recent thing, from the dreaded '09ers (I started in 2010, I’m kidding)? Or does anyone with older credentials remember it being that way? I’ve primarily found that recent Capcom games (SF4, UMVC3, SFxT) are the biggest offenders (even in the guides) but I’m having trouble nailing down older games or different communities.

N.B. I am NOT attacking one way or another of discussing frame data, I don’t really care as both have their decent reasons.

Frame data, and it’s terminology, has been around for years and years. Hell, even ST has frame data with startup, active, and recovery frames. However, it hasn’t really been as common-a-term until around SF4. Game devs and hardcore players have been using the information for years, but keep in mind that not every game freely releases frame data and hit box information. A majority of old-school players don’t even bother with that stuff, and prefer to simply play the game and learn by feel.

But SF4 also brought in a lot of players, some of which produced a lot of content for the community in both the forums and on youtube, and so really the internet has helped spread basic knowledge of frame data and it’s use.

I got no idea what the SF4 player base uses, but if a move remains inactive (invincible or vulnerable) for 3 frames and hits on the fourth, then it definitely has a 3-frame start-up. Else, a move that has absolutely no start-up (like Blanka’s balls in og SF) would be refereed to as having a start-up of one frame, even though there’s absolutely no start-up animation for it. It would make no sense.

To my understanding, a 3-frame move is active on the 3rd frame, which is why it’s a move that will punish other moves that leave your opponent at -3.

So Blanka’s upball actually does have 1-start up frame, and becomes active on the 2nd frame. The reason why there’s no start-up animation is because many start-up animations in ST look exactly like the neutral position, which is very misleading if you’re just watching the animation itself.

It’s not about a chronological order. The “X frame startup means move hits on the Xth frame” is simply the more commonly used method throughout the genre. Don’t limit yourself to SF4. Check frame data for other games, even 3D ones like Tekken and VF.

And because it is the more commonly used method, a player should assume that’s what he’s reading unless it is stated otherwise, like in the 3S frame data website.

I think the disagreements in the past few posts–all from people who I consider to be relatively knowledgeable–demonstrates that there is definitely some measure of confusion present regarding this issue.

How about we label the two styles for absolute clarity here.

I’ve been under the impression that the change-over from method A to method B was indeed a chronological break. At least for Capcom games.

The Yoga Book Hyper (ST) was the first time I’ve ever studied frame data and it definitely uses method A. T.Akiba’s SF2 site also uses this as well. As far as I know, these are some of the oldest sources of published frame data in all of fighting games. Those are both Japanese sources, for what it’s worth.

All material that I’ve ever seen for SF4, MvC3, and SFxT used method B, whether it be the SRK wiki or the Bradygames guides. That’s all been in English.

I don’t think I’ve ever looked at frame data for any other Capcom games. Same for SNK.

VF, Tekken, and Soulcalibur use method B right now, I believe. Has this always been the case?

Guilty and Blazblue both seem to use method B. Do all of the “number notation” 2D games (animeee) use method B?

Are SF2 games the only ones that have ever used method A?

I’m assuming this all matches up with how they handle frame data across the oceans.