Breakthrough - Dragons Fire a lesson to be learnt in frame data lol

[BillStickers]

Just to clarify because I see terminology getting thrown around incorrectly and it's causing confusion:

[Active Frame]

• A frame where a given move is able to hit the opponent.
• Active Frames are the total number of frames where a move is capable of hitting an opponent.

[Start-up Frames]

• The number of frames before the first Active Frame of a given move.
• This is the value shown in the move list.

[Execution Frames]

• The number of frames before and including the first Active Frame.
• This is calculated as [Start-up Frames] + 1

[Unblocking Penalty]

• It costs 1 frame to unblock before you can start an attack.

[Reversal]

• A special move, throw, or backdash performed on the first possible frame after blocking an attack.
• Negates the Unblocking Penalty.

Conclusions

• Yes, all moves are technically "one frame slower" than the frame data suggests (really it's more of a "does Christmas Day count in a days till Christmas countdown" type of situation), but it doesn't really matter since ALL moves in the game are calculated this way (so they all cancel each other out).
• @GGA Dizzy, your first post mentions that Liu Kang's b1 comes out on the 10th frame whereas Lao's spin comes out "in 7 frames." Unless mkxframedata is incorrect, Lao's spin has 7 start up frames, which means it will always come out on frame 8. Startup vs execution frames in this case are irrelevant because both fighters are subject to the same game mechanics.
• The game's block advantage numbers should be considered as accurate for reversal situations only. If the defender plans on following up with a normal, you can effectively add 1 to the first attacker's block advantage (or subtract 1 from the first defender's block advantage).

 

[AK Harold]

The wisdom of this man is boundless.

Harold. When we gonna hang and play MK? I've been practicing but your coaching is unparalleled.

You know I would be up to discuss whenever, but i have a ps4/pc ready stick. Sorry haven't gotten an xbox stick, seems most games are going to be PS4 for fighting games.

 

[Walter Fudgens]

So wait up, if 9 frame reversal normal still come out on the 10th frame, what was the patch that added an extra frame to everything for? Was this 9f normal listed as 8f before?

And does block advantage factor this in? Eg a -10 on block move factors in that normals take an extra frame to come out and thus can be punished by 10f normals, but 11f specials can also punish it cause special hreversals shave a frame ?

No, block advantage doesn't consider whether you are using a normal or a special to punish. A move that is -10 on block can only be punished by a normal that is listed with 9 frames of start up or a reversal special move listed with 10 frames of start up (or anything faster of course).

 

[AK L0rdoftheFLY]

You know I would be up to discuss whenever, but i have a ps4/pc ready stick. Sorry haven't gotten an xbox stick, seems most games are going to be PS4 for fighting games.

I have a ps pad. I'll come to you bro! Or I'll have Xeno bring his ps over.

 

[infamy23]

[Unblocking Penalty]

• It costs 1 frame to unblock before you can start an attack.

[Reversal]

• A special move, throw, or backdash performed on the first possible frame after blocking an attack.
• Negates the Unblocking Penalty.

This is not true, and I can't believe people are STILL spreading this misinformation after 8 months, when I've explained it several times on TYM.

I think I even had this discussion already on the LIU KANG forums for christ's sake.

The frame data in the game is accurate, if you know how to read it.

Moves hit on the 1st active frame.

Startup values do not include the 1st active frame.

When a reversal is performed, 1 frame of blockstun is removed, it is never added as a "penalty". The reason I know this, is because if you don't do anything after blocking an attack, I can easily count the set number of blockstun frames that each move in the game causes. Reversals are the exception, not the rule.

In-game "block advantage" values are assuming that the defender is NOT doing a reversal.

So why does b1 get beat by reversal spin? Because MB FK is not +2 against reversals, it is +1. That means your b1 will hit on frame 9.

Spin hits on frame 8.

8 < 9

End of discussion.

Can we stop this now?

 

[BillStickers]

This is not true, and I can't believe people are STILL spreading this misinformation after 8 months, when I've explained it several times on TYM.

I think I even had this discussion already on the LIU KANG forums for christ's sake.

The frame data in the game is accurate, if you know how to read it.

Moves hit on the 1st active frame.

Startup values do not include the 1st active frame.

When a reversal is performed, 1 frame of blockstun is removed, it is never added as a "penalty". The reason I know this, is because if you don't do anything after blocking an attack, I can easily count the set number of blockstun frames that each move in the game causes. Reversals are the exception, not the rule.

In-game "block advantage" values are assuming that the defender is NOT doing a reversal.

So why does b1 get beat by reversal spin? Because MB FK is not +2 against reversals, it is +1. That means your b1 will hit on frame 9.

Spin hits on frame 8.

8 < 9

End of discussion.

Can we stop this now?

Interesting idea but I'd like others to weigh in rather than shutting down discussion. Thank you for your contribution.

 

[infamy23]

There is nothing to discuss. This has been proven over and over again by Dizzy and myself, and we are both using frame by frame hardware to test these things properly, unlike Tom who just made some incorrect assumptions and then decided to post it as fact and confuse the hell out of everyone.

 

[Dizzy]

Just to clarify because I see terminology getting thrown around incorrectly and it's causing confusion:

[Active Frame]

• A frame where a given move is able to hit the opponent.
• Active Frames are the total number of frames where a move is capable of hitting an opponent.

[Start-up Frames]

• The number of frames before the first Active Frame of a given move.
• This is the value shown in the move list.

[Execution Frames]

• The number of frames before and including the first Active Frame.
• This is calculated as [Start-up Frames] + 1

[Unblocking Penalty]

• It costs 1 frame to unblock before you can start an attack.

[Reversal]

• A special move, throw, or backdash performed on the first possible frame after blocking an attack.
• Negates the Unblocking Penalty.

Conclusions

• Yes, all moves are technically "one frame slower" than the frame data suggests (really it's more of a "does Christmas Day count in a days till Christmas countdown" type of situation), but it doesn't really matter since ALL moves in the game are calculated this way (so they all cancel each other out).
• @GGA Dizzy, your first post mentions that Liu Kang's b1 comes out on the 10th frame whereas Lao's spin comes out "in 7 frames." Unless mkxframedata is incorrect, Lao's spin has 7 start up frames, which means it will always come out on frame 8. Startup vs execution frames in this case are irrelevant because both fighters are subject to the same game mechanics.
• The game's block advantage numbers should be considered as accurate for reversal situations only. If the defender plans on following up with a normal, you can effectively add 1 to the first attacker's block advantage (or subtract 1 from the first defender's block advantage).

I got tagged quoted in a few posts but I will just say here what is correct and incorrect in this post as it covers the questions other people were asking too.

• It costs 1 frame to unblock before you can start an attack. This is false

This is literally just made up and has no basis from a frame data standpoint. For example, Liu Kang's b1 is 9f start up. Hits on frame 10 at the earliest when done in neutral. Punishes moves that are -10 on block. This would not be possible if there was any unblocking penalty.

• @GGA Dizzy, your first post mentions that Liu Kang's b1 comes out on the 10th frame whereas Lao's spin comes out "in 7 frames." Unless mkxframedata is incorrect, Lao's spin has 7 start up frames, which means it will always come out on frame 8. Startup vs execution frames in this case are irrelevant because both fighters are subject to the same game mechanics.

Reversals hit 1 frame earlier than normals. If you do a Spin in neutral with Kung Lao it will hit in 8 frames, however you can punish moves that are -7 on block with spin due to a special property of reversals that removes 1 frame of blockstun.

In summary, there no "unblocking penalty". This is a bad way of thinking because it makes it sound like moves come out slower when done out of block, which is not the case. However, you can think of reversals getting a bonus. I believe that will help you get the full picture of whats going on in regards to the frame data.


Semi related, since @xarakamaka brought it up, I've considered making a thread regarding the way frame data works in the game, with examples and whatnot, but I haven't got around to it. I know it'd probably help alot but I don't want to half-ass it.

 

[Rip Torn]

There is nothing to discuss. This has been proven over and over again by Dizzy and myself, and we are both using frame by frame hardware to test these things properly, unlike Tom who just made some incorrect assumptions and then decided to post it as fact and confuse the hell out of everyone.

I agree with what you said about how the frame data is listed, but something about MKX is messed up. For example, sometimes Liu's standing 1 (8f) beats Lao's reversal spin cleanly after bicycle kick (+2). Shouldn't it only trade with spin, based on what you have said?

I'm not saying you are wrong, all I'm trying to say is that MKX is messed up and inconsistent and that there is no true way to truly understand it. This is just one of many, many inconsistencies that can be found throughout the game. I believe it has something to do with breathing hurtboxes, but that doesn't make me feel any better about it... What's your explanation? Or if @GGA Dizzy could explain?

 

[Alien Substance]

@xarakamaka, I've added two quotations to the original post, primarily so that those who may enter the thread do not have to go through two pages to find Dizzy's first reply.

 

[IrishMantis]

Just to clarify because I see terminology getting thrown around incorrectly and it's causing confusion:

[Active Frame]

• A frame where a given move is able to hit the opponent.
• Active Frames are the total number of frames where a move is capable of hitting an opponent.

[Start-up Frames]

• The number of frames before the first Active Frame of a given move.
• This is the value shown in the move list.

[Execution Frames]

• The number of frames before and including the first Active Frame.
• This is calculated as [Start-up Frames] + 1

[Unblocking Penalty]

• It costs 1 frame to unblock before you can start an attack.

[Reversal]

• A special move, throw, or backdash performed on the first possible frame after blocking an attack.
• Negates the Unblocking Penalty.

Conclusions

• Yes, all moves are technically "one frame slower" than the frame data suggests (really it's more of a "does Christmas Day count in a days till Christmas countdown" type of situation), but it doesn't really matter since ALL moves in the game are calculated this way (so they all cancel each other out).
• @GGA Dizzy, your first post mentions that Liu Kang's b1 comes out on the 10th frame whereas Lao's spin comes out "in 7 frames." Unless mkxframedata is incorrect, Lao's spin has 7 start up frames, which means it will always come out on frame 8. Startup vs execution frames in this case are irrelevant because both fighters are subject to the same game mechanics.
• The game's block advantage numbers should be considered as accurate for reversal situations only. If the defender plans on following up with a normal, you can effectively add 1 to the first attacker's block advantage (or subtract 1 from the first defender's block advantage).

I was the only person to use the term "active frame" and I used it correctly unless this was just a broad statement?

 

[skater11]

This is not true, and I can't believe people are STILL spreading this misinformation after 8 months, when I've explained it several times on TYM.

I think I even had this discussion already on the LIU KANG forums for christ's sake.

The frame data in the game is accurate, if you know how to read it.

Moves hit on the 1st active frame.

Startup values do not include the 1st active frame.

When a reversal is performed, 1 frame of blockstun is removed, it is never added as a "penalty". The reason I know this, is because if you don't do anything after blocking an attack, I can easily count the set number of blockstun frames that each move in the game causes. Reversals are the exception, not the rule.

In-game "block advantage" values are assuming that the defender is NOT doing a reversal.

So why does b1 get beat by reversal spin? Because MB FK is not +2 against reversals, it is +1. That means your b1 will hit on frame 9.

Spin hits on frame 8.

8 < 9

End of discussion.

Can we stop this now?

So instead of a 1 frame "unblocking penalty", could we say reversal special "rewards 1 frame"?

 

[BillStickers]

I got tagged quoted in a few posts but I will just say here what is correct and incorrect in this post as it covers the questions other people were asking too.

• It costs 1 frame to unblock before you can start an attack. This is false

This is literally just made up and has no basis from a frame data standpoint. For example, Liu Kang's b1 is 9f start up. Hits on frame 10 at the earliest when done in neutral. Punishes moves that are -10 on block. This would not be possible if there was any unblocking penalty.

• @GGA Dizzy, your first post mentions that Liu Kang's b1 comes out on the 10th frame whereas Lao's spin comes out "in 7 frames." Unless mkxframedata is incorrect, Lao's spin has 7 start up frames, which means it will always come out on frame 8. Startup vs execution frames in this case are irrelevant because both fighters are subject to the same game mechanics.

Reversals hit 1 frame earlier than normals. If you do a Spin in neutral with Kung Lao it will hit in 8 frames, however you can punish moves that are -7 on block with spin due to a special property of reversals that removes 1 frame of blockstun.

In summary, there no "unblocking penalty". This is a bad way of thinking because it makes it sound like moves come out slower when done out of block, which is not the case. However, you can think of reversals getting a bonus. I believe that will help you get the full picture of whats going on in regards to the frame data.


Semi related, since @xarakamaka brought it up, I've considered making a thread regarding the way frame data works in the game, with examples and whatnot, but I haven't got around to it. I know it'd probably help alot but I don't want to half-ass it.

Thank you for clarifying! Now, we need to get to the bottom of the point that @Rip Torn brought up about Liu Kang's s1 beating Lao's reversal spin after a MB bicycle kick.

We also need to get to the bottom of the point that @THTB brought up -- the patch notes state that frame 0 of an attack is now included in the start-up frames. Does this mean there was a previously unaccounted for frame in start-up, or is this supposed to mean the first active frame in the attack is now included, which would make the start-up value actually the "execution" value?

 

[Tom Brady]

Lets this finally become common knowledge...

Here is proof of a 7F normal trading. This means that if the CONFIRMED 7F normal trades that, if in fact you could not beat out Takeda's 8F poke with your B1 after MB FK, you did not do the B1 fast enough. There is no way that an 8F normal will trade/win if I am showing proof of a 7F normal TRADING.

Bottom line:
Liu Kangs B1 is 9 frames, he still locks down characters with 8F normal as their fastest attack. Characters with a 7F normal will trade. As a reversal, special's cheat 1 frame which is why 7F specials win out in this situation. There is also no such thing as 1 extra frame of guard stun for a normal. The 7F specail as a reversal will simply come out 1F faster than the 7F normal because a reversal special cheats 1 frame in how it comes out of block 1 frame faster.

 

[Dizzy]

Thank you for clarifying! Now, we need to get to the bottom of the point that @Rip Torn brought up about Liu Kang's s1 beating Lao's reversal spin after a MB bicycle kick.

We also need to get to the bottom of the point that @THTB brought up -- the patch notes state that frame 0 of an attack is now included in the start-up frames. Does this mean there was a previously unaccounted for frame in start-up, or is this supposed to mean the first active frame in the attack is now included, which would make the start-up value actually the "execution" value?

Neither really have anything to do with this thread, but...

1. I can't say I've ever tried this as it's kind of a random example, but I do know that Laos regular spin has poor range and can't actually punish moves in some situations that it should by frame data. Keep in mind frame data is just data, it does account for range at all. If this is actually happening I would not be surprised. MKX lao spin is definitely no MK9 spin when it comes to range.

I should say that I am always skeptical of anyone who says something happens "randomly" when its an interaction between two moves.

2. No idea what that means. Many of the patch notes use terms that make no sense like "auto-block" it's best to just deal with reality using testing and recording equipment rather than trying to figure out the theory behind patch notes wording.

The reality of the situation was that the start up frame data was just wrong at release and a lot of it got fixed all at once. So there was likely a bug in whatever tool was used to determine start up frame data at release, if I had to guess.

 

[infamy23]

I agree with what you said about how the frame data is listed, but something about MKX is messed up. For example, sometimes Liu's standing 1 (8f) beats Lao's reversal spin cleanly after bicycle kick (+2). Shouldn't it only trade with spin, based on what you have said?

I'm not saying you are wrong, all I'm trying to say is that MKX is messed up and inconsistent and that there is no true way to truly understand it. This is just one of many, many inconsistencies that can be found throughout the game. I believe it has something to do with breathing hurtboxes, but that doesn't make me feel any better about it... What's your explanation? Or if @GGA Dizzy could explain?

The interaction of moves in a fighting game is dependent on 2 things.

Frame data, and hitboxes.

Frame data is constant, hitboxes are dynamic.

Yes, it causes inconsistency, but I don't see how this prevents us from understanding what's going on?

There is no need to question the way frame data operates just because of a moving hitbox.

Jax can only punish Jason's F42 half the time with dash punch. Does this mean I should question whether or not Jax's dash punch is 10f? No, it means that F42 has a weird hitbox interaction with Jax's hurtbox, which causes F42 to sometimes be -9 on block.

It sucks, but it's part of fighting games. Just be thankful that it's not as bad as it is in SF4.

 

[xxFalcon Loverxx]

Lets this finally become common knowledge...

Here is proof of a 7F normal trading. This means that if the CONFIRMED 7F normal trades that, if in fact you could not beat out Takeda's 8F poke with your B1 after MB FK, you did not do the B1 fast enough. There is no way that an 8F normal will trade/win if I am showing proof of a 7F normal TRADING.

Bottom line:
Liu Kangs B1 is 9 frames, he still locks down characters with 8F normal as their fastest attack. Characters with a 7F normal will trade. As a reversal, special's cheat 1 frame which is why 7F specials win out in this situation. There is also no such thing as 1 extra frame of guard stun for a normal. The 7F specail as a reversal will simply come out 1F faster than the 7F normal because a reversal special cheats 1 frame in how it comes out of block 1 frame faster.

All hail Lord brady

 

[Rip Torn]

Neither really have anything to do with this thread, but...

1. I can't say I've ever tried this as it's kind of a random example, but I do know that Laos regular spin has poor range and can't actually punish moves in some situations that it should by frame data. Keep in mind frame data is just data, it does account for range at all. If this is actually happening I would not be surprised. MKX lao spin is definitely no MK9 spin when it comes to range.

I should say that I am always skeptical of anyone who says something happens "randomly" when its an interaction between two moves.

I was actually just trying to test if b1 was 9 frames because of what @infamy23 said about spin hitting on frame 8 and b1 hitting on frame 9. So I tested an 8 frame move to see if it would trade with spin after bicycle kick. It also happens after other strings that are +2 for Liu such as his 111 and standing 3.

When I said sometimes I was able to beat it clean, I wasn't implying it was random, just that my execution isn't always frame perfect. If an 8 frame move can win clean, you would think a 9 frame move would trade. I know about the possibility of hitbox/hurtbox issues with spin, but I was doing this point blank in the corner and it would still happen.

So I guess my (and others') theory was that the breathing animation affects the blockstun on tons of moves in the game by 1 frame + or -. It's not necessarily 'random' as there is an explanation if the theory is true. However, for a human being, it appears random, as you have no indication visually of when that extra frame might come into play.

Anyway, that's just my explanation of what I was saying. I don't really know if that's what is really happening, but that's the best explanation for the inconsistencies I've heard so far. I just wanted to know if anyone else had a better explanation.

 

[cR Xarakamaka]

I understand and watched beasts video and saying kl spin comes out on the 6th frame what he said, but why does s1 into s1 beat kl spin? 8 frame start up and it's plus 2. Spin comes out on the 6th as a reversal. That's what toms video says. So it should trade. Not win . Is s1 , 7 frames??? Someone explain please please @GGA Dizzy @Tom Brady thanks bros your a big help

 

[Tom Brady]

I understand and watched beasts video and saying kl spin comes out on the 6th frame what he said, but why does s1 into s1 beat kl spin? 8 frame start up and it's plus 2. Spin comes out on the 6th as a reversal. That's what toms video says. So it should trade. Not win . Is s1 , 7 frames??? Someone explain please please @GGA Dizzy @Tom Brady thanks bros your a big help

Spin probably has MK9 like properties lol. In MK9 there was something known as player 1 adv. This meant that in a game with no trades, player 1 would win every situation where a trade would happen. This however did not apply for KL's spin. Regardless on if you were player 1 or 2, KL's spin wins out. In the situation you brought up, the spin is winning when it should trade. The only way the reversal spin would lose in this situation is if the follow up attack comes out faster then the spin.

 

[xxFalcon Loverxx]

Spin probably has MK9 like properties lol. In MK9 there was something known as player 1 adv. This meant that in a game with no trades, player 1 would win every situation where a trade would happen. This however did not apply for KL's spin. Regardless on if you were player 1 or 2, KL's spin wins out. In the situation you brought up, the spin is winning when it should trade. The only way the reversal spin would lose in this situation is if the follow up attack comes out faster then the spin.

Lord Brady strikes again!

 

[Rip Torn]

Spin probably has MK9 like properties lol. In MK9 there was something known as player 1 adv. This meant that in a game with no trades, player 1 would win every situation where a trade would happen. This however did not apply for KL's spin. Regardless on if you were player 1 or 2, KL's spin wins out. In the situation you brought up, the spin is winning when it should trade. The only way the reversal spin would lose in this situation is if the follow up attack comes out faster then the spin.

No, the spin is losing when it should trade. That's what is so mystifying about it.

 

[Tom Brady]

No, the spin is losing when it should trade. That's what is so mystifying about it.

ahhh.. I misread what he was saying. hmmm... i'll do some testing. But I do for a fact know that reversal specials cheat 1 frame from block stun.

 

[Dizzy]

I understand and watched beasts video and saying kl spin comes out on the 6th frame what he said, but why does s1 into s1 beat kl spin? 8 frame start up and it's plus 2. Spin comes out on the 6th as a reversal. That's what toms video says. So it should trade. Not win . Is s1 , 7 frames??? Someone explain please please @GGA Dizzy @Tom Brady thanks bros your a big help

Why are you saying 6 frames? It's 7...lol. look at the move list.

 

[cR Xarakamaka]

Why are you saying 6 frames? It's 7...lol. look at the move list.

Im only going by what I'm being taught , look at Tom beasts video he posted dizzy

2:35

Not me who said it , I'm just trying to understand what's going on . But I look forward to your thread dude