How to Read, Understand and Calculate Frame Data in MKX

[Tharr]

Either I get this wrong or something's off. Balanced Kenshi's 1,1 is 24 cancel, 0 block adv, so 24 cancel block adv.
Rising Karma(db1) is 16f startup, so plenty of time for blockstring? Nope, AI dashes backwards / armors out.
Tele-flurry(bf3) is 13f startup, AI doesn't dash or armor out on reversal. How does 24 cancel block adv not work on 16f but works on 13f?

Also, Scorp's 1,2,3 isn't backdashed or armored out of by AI, and it's the example in the main post. So it is a blockstring now?

@Eddy Wang

 

[Eddy Wang]

Either I get this wrong or something's off. Balanced Kenshi's 1,1 is 24 cancel, 0 block adv, so 24 cancel block adv.
Rising Karma(db1) is 16f startup, so plenty of time for blockstring? Nope, AI dashes backwards / armors out.
Tele-flurry(bf3) is 13f startup, AI doesn't dash or armor out on reversal. How does 24 cancel block adv not work on 16f but works on 13f?

Also, Scorp's 1,2,3 isn't backdashed or armored out of by AI, and it's the example in the main post. So it is a blockstring now?

@Eddy Wang

wrong in game frame data

 

[KimCheeze]

The frame data for the combo section corresponds to each hit in the combo. It was also like this in Injustice.

For example, Scorpion 123 string will initially be broken down into 12, and then it'll have 123 under it.

For the 12 string, it'll say 6f startup, and is referring to the 2. For the 123 string it'll say 8f startup, and is referring to the 3.

If you go on Injustice and look up Nightwing's 1F2 string, it'll say 15f startup. This is referring to the F2 in the string. Standing 1 is 11f startup so the whole string has a startup of 26f.

@Eddy Wang could we update the OP with this information? honestly this is like enlightenment when it comes to the frame data information in MKX (especially for those of us who did not play Injustice nor MK9)

 

[PsychoNoob]

Shouldn't mileena's b2,2,3+4 not be a true blockstring? I set the cpu mileena dummy to always block and reverse with ex roll while I do b2,2,3+4, but the 3+4 doesn't get interrupted. B22 has a cancel block adv of 10f, and the 3+4 in the b22,3+4 is 19f of startup. What am I not getting here?

 

[Kombat_Killer908]

I'm confused about something. I'm trying to improve my game and trying to understand frame data. I use Shinnok, his 1 starts up in 9 frames, so why does,1 1 start up in 13, and 1 1 2 starts up in 17

 

[PsychoNoob]

I'm confused about something. I'm trying to improve my game and trying to understand frame data. I use Shinnok, his 1 starts up in 9 frames, so why does,1 1 start up in 13, and 1 1 2 starts up in 17

the 13 frames on the 1,1 string is regarding the startup frames of the second hit in 1,1. The 17 frames of startup is regarding the last hit in 112. The first hit in all those strings still come out in 9 frames they're just telling you the startup of the last hit in the string.

 

[Kombat_Killer908]

the 13 frames on the 1,1 string is regarding the startup frames of the second hit in 1,1. The 17 frames of startup is regarding the last hit in 112. The first hit in all those strings still come out in 9 frames they're just telling you the startup of the last hit in the string.

oh ok thank you, sorry for the noobish question haha

 

[huh]

Guys, I was calculating cancel advantages and having trouble since some moves, according to the formula cancel+hit advantage = cancel advantage on hit, and cancel+block advantage = cancel advantage on block, were not consistent. Block strings were not block strings, and some combos that theoretically could not be cancelled into specials were in fact cancelable.

Add also the fact that some cancel numbers ingame are higher than the recovery frames, which didn't make any sense to me. How can you have a 16 frame cancel advantage in a move that has 11 recovery frames? You would cancel into the active frames!! So, there is something wrong here.

Up to now, I was considering those inconsistencies were from erroneous ingame frame data. But then I found this post from Myokymia, and he actually makes sense:

http://steamcommunity.com/sharedfiles/filedetails/?id=428638611&insideModal=1

Basically, he says that the cancel data that the game shows is not the cancel advantage. Is actually the number of frames it takes, since the startup of the move, to make the special come out.

So, for example, let's pick Reptile's 1 (front punch) frame data:

startup: 10
active: 1
recover: 11
block adv: 2
hit adv: 11
cancel: 16

The math would be:

Total frames of the move, from startup up to the end of recovery: 10 (startup) + 1 (active) + 11 (recovery) = 22
Cancel = frames BEFORE the special comes out = 16

So, to calculate the real Cancel advantage, you have to do: 22 total frames - 16 cancel frames = 6 frames cancel advantage.

Cancel Advantage on hit = 11 (hit adv.) + 6 (calculated cancel adv.) = 17
Cancel Advantage on block = 2 (block adv.) + 6 (calculated cancel adv.) = 8

The old math would be :
Cancel Advantage on hit: 11+16 = 27
Cancel Advantage on block = 2+16 = 18

So, which one is correct? Let's pick the klaw pounce special, which has 24 startup frames, to check it up.

According to the old math, 1~klaw pounce should combo, since the cancel adv. on hit is 27.
According to the new method, 1~klaw pounce doesn't combo, since the cancel adv. on hit is only 17

Guess what? 1~klaw pounce doesn't combo.

So, that new method of reading the cancel frame data is much more consistent with the game data and a lot of frame windows now make sense.

@Eddy Wang I think you should see this and perhaps change the OP!

 

[kencheese]

Guys, I was calculating cancel advantages and having trouble since some moves, according to the formula cancel+hit advantage = cancel advantage on hit, and cancel+block advantage = cancel advantage on block, were not consistent. Block strings were not block strings, and some combos that theoretically could not be cancelled into specials were in fact cancelable.

Add also the facto that some cancel numbers in game are higher than the recovery frames, which didn't make any sense to me. How can you have a 16 frame cancel advantage in a move that has 11 recovery frames? You would cancel into the active frames!! So, there is something wrong here.

Up to now, I was considering those inconsistencies were from erroneous in game frame data. But then I found this post from Myokymia, and he actually makes sense:

http://steamcommunity.com/sharedfiles/filedetails/?id=428638611&insideModal=1

Basically, he says that the cancel data that the game shows is not the cancel advantage. Is actually the number of frames it takes, since the startup of the move, to make the special come out.

So, for example, let's pick Reptile's 1 (front punch) frame data:

startup: 10
active: 1
recover: 11
block adv: 2
hit adv: 11
cancel: 16

The math would be:

Total frames of the move, from startup up to the end of recovery: 10 (startup) + 1 (active) + 11 (recovery) = 22
Cancel = frames before the special comes out = 16

So, to calculate the real Cancel advantage, you have to do: 22 total frames - 16 cancel frames = 6 frames cancel advantage.

Cancel Advantage on hit = 11 (hit adv.) + 6 (calculated cancel adv.) = 17
Cancel Advantage on block = 2 (block adv.) + 6 (calculated cancel adv.) = 8

The old math would be :
Cancel Advantage on hit: 11+16 = 27
Cancel Advantage on block = 2+16 = 18

So, which one is correct? Let's pick the law pounce special, which has 24 startup frames

According to the old math, 1~klaw pounce should combo, since the cancel adv. on hit is 27.
According to the new method, 1~klaw pounce doesn't combo, since the cancel adv. on hit is only 17

Guess what? 1~klaw pounce doesn't combo.

So, that new method of reading the cancel frame data is very consistent with the game data and a lot of frame windows now make sense.

@Eddy Wang I think you should see this and perhaps change the OP!

thank you someone finally gets it, wang and steam comm. were completely wrong, i called wang out on this and he would not respond to me, i am just a noob to this and to the game, but i did both of the old calcs and none of it worked, almost just the opposite, i did some of my own and came up with what you did, just never posted it up, what they did was pick a character to meet their calculations, and none of it worked on my figther, good work, nice and ballsies of you, keep it up scott r

 

[huh]

thank you someone finally gets it, wang and steam comm. were completely wrong, i called wang out on this and he would not respond to me, i am just a noob to this and to the game, but i did both of the old calcs and none of it worked, almost just the opposite, i did some of my own and came up with what you did, just never posted it up, what they did was pick a character to meet their calculations, and none of it worked on my figther, good work, nice and ballsies of you, keep it up scott r

Thanks man. Credit goes to that Myokymia guy.

Nice to know you reached the same conclusions.

If you think of that Cancel data as "the startup frames of the cancel" or "cancel startup" it will make even more sense.
It is how long the cancel takes to happen, within the move.

 

[ChaosTheory]

@huh

So if I'm reading this correctly, that makes the original post in this thread a definitive error, right?

 

[huh]

@huh

So if I'm reading this correctly, that makes the original post in this thread a definitive error, right?

Regarding the cancel advantage calculation yes, I think so. The other info is all correct, except that.

In fact, if it is so, that is an error that 99% of this community, me included, were making so far. The cancel numbers in other games are calculated like the OP said and because of that it is very misleading. Add to that the fact that the frame data info was incorrect in the game and there you have another factor to explain why this miscalculation persisted up to now.

For what I've seen, unless someone show me something that makes more sense, I pretty much believe the cancel advantage is calculated the way I've described, and not the way it is in the OP.

@Eddy Wang Since it is a well known forum, and this topic is one of the top listings in a google search for frame data in MKX, I think we should look into that and if confirmed, correct the info in the OP. It could benefit the community to be sure we have the right info there. Please, review the posts above.

@CrimsonShadow @STORMS @THTB @A F0xy Grampa @Pig Of The Hut @GGA 16 Bit @GGA pimpimjim @Tom Brady @colt and any other that understands well this:

Could you please check my post above regarding the cancel advantage calculation? We need to clarify that in order to help the community. If the OP is wrong, it could mislead a lot of players. And if it is right, it would be good if someone explain why some of the moves, like the example I've presented with reptile's 1, are not consistent with that.

 

[Eddy Wang]

@huh, i will give it a go and confirm your data, great find.

 

[Eddy Wang]

@huh, Cancel section is corrected, however i would prefer to update with a new guide later on, there is so much that changed withing the game, that a more accurate guide now would make sense.

Although i know exactly what changed currently i didn't found the time to sit down and complete my 2.0 guide, thanks a lot for this.

 

[huh]

@Eddy Wang Keep up the good work man. The 2.0 guide will be great, and I'm glad I was able to help somehow.

 

[kencheese]

@Eddy Wang Keep up the good work man. The 2.0 guide will be great, and I'm glad I was able to help somehow.

will send you some info on cance ladvantage on block for punishing abilities when coming out of block, will send later and please check my calculations to be sure it is right, scott

 

[kencheese]

Thanks man. Credit goes to that Myokymia guy.

Nice to know you reached the same conclusions.

If you think of that Cancel data as "the startup frames of the cancel" or "cancel startup" it will make even more sense.
It is how long the cancel takes to happen, within the move.

okay, for cancel on block adv., lets say i hit my opponent with a move, to determine if the opponent can come out of block and punish me is, take the start up frames+active frames+recovery frames=total move frames of my move that hit him , now take the total frames and subtract the block adv frames to get the cancel block adv frames, and if the opponent blocks my move and comes out of block at the very first possible frame to come out block(no hesitation=hesitation frames=wasted frames)and launches a punish of a special move, the start up frame of the special move must be equal to or less than the frame number of the calculate cancel block adv frames to punish me(it takes no frames to come out of block with a special move), if he picks a normal move(basic move or a basic kombo move, both require one frame to come out of block) he needs to pick a normal move that is one frame less on start up(assuming he came out of block at the very first possible frame to unblock) than my moves calculate frame number of cancel block advantage, and i right or just misguided, practice this in custom CPU and tested many AI reversals for many character to my character and it works if iam wrong please correct thanks for your time scott r, this is the reason the single player mode of fighting the CPU on hard and very hard is so difficult, not only is the CPU reading your inputs to counter you, but when you launch a move and the CPU blocks, the CPU is coming out of block with no hesitation like if feel it does on easy and very easy, and to me medium is more of a true human reaction all around, i use hard and very hard to practice blocking and reversals and dont get carried away with trying to beat the CPU, and when i go back to medium, medium seems to be in slow motion, and with time i will scale it up to hard and then very hard, the CPU is not cheating like people have said, it is just reading your moves faster and coming out of block exactly when it can to punish you, so as to sharpen your skills and prepare you for really competitive online play with experienced players, i have read that top pros practice this way, hope this is right scott

 

[huh]

Hey @kencheese I think there is a little mistake in what you said. Actually the formula is:

(Startup + Active + Recover) - Cancel + Block Advantage. That will give your Cancel Advantage on Block.
After getting it, you have to see if there is any special move that has less startup frames than that. If so, you can input that special and it will link without holes and can not be punished.

-------------

Just an observation, because from what you wrote I don't know if you are talking about this:

Cancel Advantage on Block only applies if you link a special move after that. If you don't input a special afterwards, it is not a cancel anymore and you have to make another calculation:

Just check the Advantage on block frames. Now, see the startup of the next move you want to input.

Do this: Startup frames from the next move - advantage on block from the first move

The result will be the amount of frames your opponent has to punish you. If he has a move with less startup frames than that, he will hit you before you hit him.

Regarding the normals vs. special 1 frame difference on reversals, I really don't know. Several people are saying this, that a normal has 1 extra frame before startup. But I really don't know if that is true.

@Eddy Wang do you know?

 

[Eddy Wang]

Hey @kencheese I think there is a little mistake in what you said. Actually the formula is:

(Startup + Active + Recover) - Cancel + Block Advantage. That will give your Cancel Advantage on Block.
After getting it, you have to see if there is any special move that has less startup frames than that. If so, you can input that special and it will link without holes and can not be punished.

-------------

Just an observation, because from what you wrote I don't know if you are talking about this:

Cancel Advantage on Block only applies if you link a special move after that. If you don't input a special afterwards, it is not a cancel anymore and you have to make another calculation:

Just check the Advantage on block frames. Now, see the startup of the next move you want to input.

Do this: Startup frames from the next move - advantage on block from the first move

The result will be the amount of frames your opponent has to punish you. If he has a move with less startup frames than that, he will hit you before you hit him.

Regarding the normals vs. special 1 frame difference on reversals, I really don't know. Several people are saying this, that a normal has 1 extra frame before startup. But I really don't know if that is true.

@Eddy Wang do you know?

Unblocking Costs 1 frame so punishing without being a reversal has to be done with a move that is 2 frames faster than the negative blockstun.

The game before allowed the attacker to block at -1 it doesn't appear to be the case anymore.
Every move in the game was added 1f in the startup.

Well that is the problem with patches, it outdates a lot of things, and i think this guide is outdated already, i'm actually reluctant wondering if should let it open until i get my time to work on a updated one, or if i should ask for closing.

 

[kencheese]

Hey @kencheese I think there is a little mistake in what you said. Actually the formula is:

(Startup + Active + Recover) - Cancel + Block Advantage. That will give your Cancel Advantage on Block.
After getting it, you have to see if there is any special move that has less startup frames than that. If so, you can input that special and it will link without holes and can not be punished.

-------------

Just an observation, because from what you wrote I don't know if you are talking about this:

Cancel Advantage on Block only applies if you link a special move after that. If you don't input a special afterwards, it is not a cancel anymore and you have to make another calculation:

Just check the Advantage on block frames. Now, see the startup of the next move you want to input.

Do this: Startup frames from the next move - advantage on block from the first move

The result will be the amount of frames your opponent has to punish you. If he has a move with less startup frames than that, he will hit you before you hit him.
you are absoutley correct, i was typing to fast and trying to cram to much into 5 minutes, i meant to type su+a+r-c+ba=cancel on block adv
Regarding the normals vs. special 1 frame difference on reversals, I really don't know. Several people are saying this, that a normal has 1 extra frame before startup. But I really don't know if that is true.

@Eddy Wang do you know?

Hey @kencheese I think there is a little mistake in what you said. Actually the formula is:

(Startup + Active + Recover) - Cancel + Block Advantage. That will give your Cancel Advantage on Block.
After getting it, you have to see if there is any special move that has less startup frames than that. If so, you can input that special and it will link without holes and can not be punished.

-------------

Just an observation, because from what you wrote I don't know if you are talking about this:

Cancel Advantage on Block only applies if you link a special move after that. If you don't input a special afterwards, it is not a cancel anymore and you have to make another calculation:

Just check the Advantage on block frames. Now, see the startup of the next move you want to input.

Do this: Startup frames from the next move - advantage on block from the first move

The result will be the amount of frames your opponent has to punish you. If he has a move with less startup frames than that, he will hit you before you hit him.

Regarding the normals vs. special 1 frame difference on reversals, I really don't know. Several people are saying this, that a normal has 1 extra frame before startup. But I really don't know if that is true.

@Eddy Wang do you know?

you are correct i made a mistake, trying to type to fast....cancel on block adv=su + a+r-c+block adv, and this number must be higher than the start up(su) of your move your are canceling into, what is your opinion of liu kang, iam new to mk and already love this character to play with as my main figther.

 

[ChaosTheory]

I was messing with Necromancer Shinnok block-strings today. Are there still a lot of errors in the frame-data that people are noticing?

Long story short... The AI armors through the final hit on 1,1,2,3. Going by the formula, that makes sense because there's a 5-frame gap. However, there is a 2-frame gap between second and third hits (1,1,2,3) but no interruption.

So is the frame-data wrong or am I misunderstanding something about the gaps/reversals/armor?

 

[Algag85]

I've seen this on multiple characters I believe, but I'll use Mileena as an example. Her F1 Sai-Fan in piercing is 19 seconds on Start-up, but her kombo Double Impale, F1, 2+4, 1+3, is listed as a 1 second startup. How are they supposedly different start up times? She has other kombos that start with F1 and they all have different start up times. There is even a F1, 4 that is 22 start-up, and a F1, 4, 3 that is a 24 Start-up. Is this all calculated for and adjusted when you input each command. So with the F1, 4, 3, when I complete the final input of 3 does it speed up the start-up by 2 frames? But this doesn't make sense, because I can hit the 3 after F1 has already completed and 3 has begun. Any explanations?

 

[Tharr]

I've seen this on multiple characters I believe, but I'll use Mileena as an example. Her F1 Sai-Fan in piercing is 19 seconds on Start-up, but her kombo Double Impale, F1, 2+4, 1+3, is listed as a 1 second startup.

Because the startup in combos is for the last hit in the combo listed.
f1 = 19f startup
f1,4 = 19f startup for f1 + 22f startup for 4
f1,4,3 = 19f + 22f + 24f startup for 3.

 

[huh]

Guys, after so many changes, I'm kind of lost here.

Up to today I always thought a -8 on block move could only be punished by a 7 frame special and a 6 frame normal.

But I've heard that there is no more 1 frame to unblock, and thee just frame punish is possible, and so a punish with a normal or with a reversal will come out in the same frame.

Well, just a simple question then:

So, a -8 on block move can be punished with a 8 frame special and a 8 frame normal?

 

[huh]

bump