Menu
By:
Filters
What's new

MK9 Frame Data Project

D

DrDogg

Kombatant
As most of you know, frame data is extremely valuable to competitive players. With the lack of MK frame data, Lomyn started to calculate his own. I have joined in on this venture and we'll be doing all of the characters. We'll start with execution frames (these can be done without the aid of another human), then move on to hit and block frame data.

This will be a long process, probably taking a month or more to get everything completely done. So I thought I'd make this thread so everyone can gauge our progress, see which characters are coming next, and to ensure that we aren't doing work that other people have already started.

If you planned to do frame data, or have already done some, please post here so we don't do any extra work. If you'd like to assist with frame data analysis and have the ability to record at 60 fps and play back frame by frame at 60 fps please post up.

I ask that you do not request specific characters. All of the characters will be done, but Lomyn and I are going in the order we prefer.

Here is the character list thus far. As each character is completed, their frame data will be found in the character forums:

Execution Frames

Lomyn
- Kitana: http://testyourmight.com/forum/showthread.php?4391-Kitana-Frame-Data
- Ermac: http://testyourmight.com/forum/showthread.php?4641-Ermac-Frame-Data
- Kung Lao: http://testyourmight.com/forum/showthread.php?4892-Kung-Lao-Frame-Data

DrDogg
- Cage: http://testyourmight.com/forum/showthread.php?4952-Johnny-Cage-Frame-Data
- Sonya: http://testyourmight.com/forum/showthread.php?4954-Sonya-Frame-Data
- Raiden: http://testyourmight.com/forum/showthread.php?4955-Raiden-Frame-Data
- Stryker: http://testyourmight.com/forum/showthread.php?4953-Stryker-Frame-Data
- Jax: http://testyourmight.com/forum/showthread.php?5191-Jax-Frame-Data
- Sindel: http://testyourmight.com/forum/showthread.php?5195-Sindel-Frame-Data
- Liu Kang: http://testyourmight.com/forum/showthread.php?5193-Liu-Kang-Frame-Data
- Nightwolf: http://testyourmight.com/forum/showthread.php?7295-Nightwolf-Frame-Data
- Sheeva: http://testyourmight.com/forum/showthread.php?7294-Sheeva-Frame-Data
- Kabal: http://testyourmight.com/forum/showthread.php?6444-Kabal-Frame-Data&p=84674#post84674
- Reptile: http://testyourmight.com/forum/showthread.php?7751-Reptile-Frame-Data
- Scorpion: http://testyourmight.com/forum/showthread.php?7784-Scorpion-Frame-Data
- Sub-Zero: http://testyourmight.com/forum/showthread.php?7785-Sub-Zero-Frame-Data
- Kenshi: http://testyourmight.com/forum/showthread.php?7799-Kenshi-Frame-Data

Sprint
- Cyrax: http://testyourmight.com/forum/showthread.php?4992-Cyrax-Frame-Data

TheSaBiie
- Smoke: http://testyourmight.com/forum/showthread.php?8943-Smoke-Frame-Data
- Kano: http://testyourmight.com/forum/showthread.php?9177-Kano-Frame-Data

XXZ
- Skarlet: http://testyourmight.com/forum/showthread.php?11958-Skarlet-Frame-Data
- Mileena: http://testyourmight.com/forum/showthread.php?11959-Mileena-Frame-Data
- Jade: http://testyourmight.com/forum/showthread.php?11973-Jade-Frame-Data
- Noob: http://testyourmight.com/forum/showthread.php?9291-Noob-Saibot-Frame-Data
- Shang Tsung: http://testyourmight.com/forum/showthread.php?11991-Shang-Tsung-Frame-Data
- Sektor: http://testyourmight.com/forum/showthread.php?11989-Sektor-Frame-Data
- Cyber Sub-Zero: http://testyourmight.com/forum/showthread.php?11986-Cyber-Sub-Zero-Frame-Data
- Quan Chi: http://testyourmight.com/forum/showthread.php?11985-Quan-Chi-Frame-Data

Riu48
- Baraka: http://testyourmight.com/forum/showthread.php?11972-baraka-frame-data

Hit/Block Frames

Pending...

Notes of Interest

Fastest Attack: Kung Lao's Spin and X-Ray at 6 frames (W.T.F?!). Both Sonya and Kung Lao's Dive Kicks are estimated at 6 frames as well.

Throws: All throws seem to execute in 10 frames. I'm going to see if I can calculate the escape window, as well as how long it takes to release block. I'm guessing it's less than a 10 frame break window. Maybe around 6 or 7 frames.

Jax: His normal Ground Pound is 41 frames, one of the slowest specials in the game. His EX Ground Pound is even slower at 48 frames, which is one of the slowest EX specials in the game. Thanks NRS... >_>

- Jumping Forward/Back Punch seems to be 7 frames for execution.

- Neutral Jump Punch seems to be 9 frames for execution. This seems to be the first active frame, but I've seen it hit on the 10th frame as well. I'm assuming I either started the punch early or the opponent was somehow lower.

Both jump forward and neutral jump punch seem to be universal. I have not tested every character, but I tested several and they were all the same speed.
 
E

evansgambit

Guardian of Outworld
Possibly have to include version dates

Hey DrDogg, great thread. Something like this is just so valuable.

Given that netherrealm, are going to patch this game on the fly with various updates.
I would suggest that all frame data info be dated. With these we could reference what moves have been frame updated.

Come to think of it, if they keep updating the game, when would it be "officially finished" and ready to go.

Cheers,
 
D

DrDogg

Kombatant
Updated the OP with links to Cage, Raiden, Sonya and Stryker.

evansgambit said:
Hey DrDogg, great thread. Something like this is just so valuable.

Given that netherrealm, are going to patch this game on the fly with various updates.
I would suggest that all frame data info be dated. With these we could reference what moves have been frame updated.

Come to think of it, if they keep updating the game, when would it be "officially finished" and ready to go.

Cheers,
Click to expand...
NRS has stated that we'll get patch notes when anything is changed in the game. Hopefully that will make it easier for us to quickly check existing frame data to determine what's changed.

As far as tracking the dates, the forum kind of does that for us. Anytime a post is edited, the time and date of the edit are listed at the bottom of the post. Anything of interest that's edited will most certainly be noted in the OP of each character's frame data thread.
 
THTB

THTB

Arez | Booya | Riu48 - Rest Easy, Friends
Good stuff, Dogg and Lomyn.

Interesting tidbit on the basic normals...none of them seem faster than 7f, but the speeds tend to be all over the place. Some normals go as high as 18f...lol.
 
Lomyn

Lomyn

Snoctopus
Damn, DrDogg is a workhorse, he got done in one weekend more than I could in two weeks! Oh well. I should have Mileena and Noob up before midnight. If there is anyone in the Minneapolis area who wouldn't mind meeting up to test some of this with me, I'd really appreciate it.
 
D

DrDogg

Kombatant
Lomyn said:
Damn, DrDogg is a workhorse, he got done in one weekend more than I could in two weeks! Oh well. I should have Mileena and Noob up before midnight. If there is anyone in the Minneapolis area who wouldn't mind meeting up to test some of this with me, I'd really appreciate it.
Click to expand...
To be fair, you're being more thorough than I am with formatting and stuff. I'm just trying to blaze through it, then once the hard part is done, I'll make it look pretty.

I won't have any time to work on this today or tomorrow, but I should get a couple more characters done on Wed. I'm shooting to have at least another 6 done by Friday night, but we'll see how the week goes. I have MK tournaments Sat and Sun, so I won't get much done this weekend.
 
S

Sprint

Noob
60fps video.

1. First frame: The frame where the limb glows red, or the frame before the button press is displayed.
2. First active frame: The frame before the damage display appears, or the frame where the opponent starts to recoil.
3. First cancelled frame: The first frame of a special move when cancelled into, identified by the red limb. Can also be located by finding the frame on which the special hits and subtracting its startup.
4. First fully recovered frame: Hold up and identify the first frame of the jumping animation. Sometimes hard to tell due to animation blending, so it's easier to use a reference frame later in the jump and subtract the known jump duration up to that point.
5. Frame advantage: Hold up with both players and measure the difference between the jumping animations. In my case I controlled P2 with my feet to speed things up, but if you want you can pause the moment an attack hits and start holding up on both controllers during the pause menu.

Startup = First active frame - First frame
Execution = Startup + 1
Cancelled duration = First cancelled frame - First frame
Duration = First fully recovered frame - First frame
Frame advantage = As described above.
Cancel advantage = As above, but cancel into a whiffed special move and add its duration to the result.

For cases where the first frame of a chained move can't easily be identified, go to the first frame of the previous move and add its cancelled duration.
 
Lomyn

Lomyn

Snoctopus
Sprint said:
60fps video.

1. First frame: The frame where the limb glows red, or the frame before the button press is displayed.
2. First active frame: The frame before the damage display appears, or the frame where the opponent starts to recoil.
3. First cancelled frame: The first frame of a special move when cancelled into, identified by the red limb. Can also be located by finding the frame on which the special hits and subtracting its startup.
4. First fully recovered frame: Hold up and identify the first frame of the jumping animation. Sometimes hard to tell due to animation blending, so it's easier to use a reference frame later in the jump and subtract the known jump duration up to that point.
5. Frame advantage: Hold up with both players and measure the difference between the jumping animations. In my case I controlled P2 with my feet to speed things up, but if you want you can pause the moment an attack hits and start holding up on both controllers during the pause menu.

Startup = First active frame - First frame
Execution = Startup + 1
Cancelled duration = First cancelled frame - First frame
Duration = First fully recovered frame - First frame
Frame advantage = As described above.
Cancel advantage = As above, but cancel into a whiffed special move and add its duration to the result.

For cases where the first frame of a chained move can't easily be identified, go to the first frame of the previous move and add its cancelled duration.
Click to expand...
The holding up method doesn't produce accurate results, at least not as far as hit advantage is concerned. You can definitely block and attack faster than you can jump out in a lot of cases.
 
S

Sprint

Noob
You can pretty much always block. Even if the opponent is at a 3 billion frame advantage on hit, you can block. Hell, you can even block while knocked flat on the floor.

The moment you enter a neutral state, the opponent can block.
The moment you connect with the last hit in a Kombo chain, the opponent can block.

The only exceptions to this rule are when the opponent is airborne or in a disabled/frozen state (freeze, spin, net, etc.)

An easy example that comes to mind is jump punches. With Cyrax you can do Jump punch -> f+BP, which combos even though it has a startup of almost half a second.
...but if you input his 6 frame crouch jab even a single frame after landing from the Jump punch, the opponent can just block it because you entered a neutral state, even though it's still connecting over 20 frames earlier than the aforementioned f+BP.


As far as I'm aware, there is nothing that allows a player to attack before they can move. Both abilities come back simultaneously.
 
D

DrDogg

Kombatant
It sounds like your execution frame data should be accurate, but I also have to question the hit/block data. The problem is that we don't know the frames for jumping or blocking. I don't believe it's instant in either case, despite what the animation indicates.

Also, I'm unsure what the benefit of knowing duration and cancel data would be. What are you using it for?
 
S

Sprint

Noob
DrDogg said:
It sounds like your execution frame data should be accurate, but I also have to question the hit/block data. The problem is that we don't know the frames for jumping or blocking. I don't believe it's instant in either case, despite what the animation indicates.

Also, I'm unsure what the benefit of knowing duration and cancel data would be. What are you using it for?
Click to expand...
Speak for yourself.

Blocking is instant (verified by pressing block exactly 10f after a inputting a move with 10f startup).

Cyrax's neutral jump has a 41 frame total duration (including startup and recovery).
The first frame can be hard to spot after certain other animations, so I used the easily identifiable 41st frame as a reference point and subtracted 40 to verify the location of the first frame in each case.

Cancel advantage and total duration can be used to calculate the advantage of a connected move cancelled into a whiffed move. e.g. Cyrax's d+BK cancelled into Anti-Air (d,f,FP):
+29 - 35 = -6f on block
+34 - 35 = -1f on hit
(This gets ridiculous with Sindel, since levitate into instant landing takes 19f total)

Cancelled duration allows the total startup and duration of chains to be calculated without having to individually list everything.
e.g. Cyrax FP,FP,FP:
13 + 15 = 28f (2nd punch hits)
13 + 24 + 17 = 54f (3rd punch hits)
13 + 24 + 47 = 84f (total duration)
 
D

DrDogg

Kombatant
Sprint said:
Speak for yourself.

Blocking is instant (verified by pressing block exactly 10f after a inputting a move with 10f startup).

Cyrax's neutral jump has a 41 frame total duration (including startup and recovery).
The first frame can be hard to spot after certain other animations, so I used the easily identifiable 41st frame as a reference point and subtracted 40 to verify the location of the first frame in each case.

Cancel advantage and total duration can be used to calculate the advantage of a connected move cancelled into a whiffed move. e.g. Cyrax's d+BK cancelled into Anti-Air (d,f,FP):
+29 - 35 = -6f on block
+34 - 35 = -1f on hit
(This gets ridiculous with Sindel, since levitate into instant landing takes 19f total)

Cancelled duration allows the total startup and duration of chains to be calculated without having to individually list everything.
e.g. Cyrax FP,FP,FP:
13 + 15 = 28f (2nd punch hits)
13 + 24 + 17 = 54f (3rd punch hits)
13 + 24 + 47 = 84f (total duration)
Click to expand...
Well I know in at least some cases blocking is not instant.

Also, I understand the method you used to calculate the canceled and total duration frames, I'm just not understanding the benefit of doing so. How does knowing how many frames it takes to execute the second punch in Cyrax's combo benefit you?

I'm not saying you're wrong and there's no benefit, I'm just trying to figure out what that benefit is because I'm not seeing it.
 
S

Sprint

Noob
DrDogg said:
Well I know in at least some cases blocking is not instant.
Click to expand...
List them. Don't waste my time with rhetoric.

DrDogg said:
Also, I understand the method you used to calculate the canceled and total duration frames, I'm just not understanding the benefit of doing so. How does knowing how many frames it takes to execute the second punch in Cyrax's combo benefit you?

I'm not saying you're wrong and there's no benefit, I'm just trying to figure out what that benefit is because I'm not seeing it.
Click to expand...
The first hit isn't always guaranteed to connect. In this case it's not particularly useful, but there may be situations where whiffing a move on purpose is practical.
Like when trying to punish a move with a low hitbox, it may be possible to connect for more damage with a string that starts with a whiffed crouchable move if the second hit comes out quickly enough.

Cyrax himself doesn't really benefit from this particular example as far as I can tell, but I'm not going to just pass my own judgement on this crap and leave out data that someone else may want to know.
 
Lomyn

Lomyn

Snoctopus
According to the hit advantage information that I tested with Kitana, according to the jump method d+1 should be -3 on hit, which would mean that after throwing it out she would lose 100% to an opposing Kitana's d+1 if they both performed that attack the first frame that they were able to act. I lack the means to test that myself, but I can say that if properly timed the Kitana that hit with the d+1 can absolutely do another d+1 before they would be allowed to jump by holding up. That's why I removed the hit data from my Kitana OP, something seemed off.

It could also simply be that the testing method requires up to be pressed at the first frame where you are out of hitstun/animation, similar to doing Seth's toe taps in the corner in SSFIV.

I think what we need to do is determine for certain exactly at which point after being hit/blocked the opponent is allowed to ATTACK. Regardless of any animation fluff that happens to make the game's hits look pretty, if an opponent can attack before their hit animation wears off, that is the point of reference for hit advantage/disadvantage.
 
S

Sprint

Noob
There appears to be a 1 frame transition between crouching and standing that I overlooked.

Kitana's d+1 is -1 on hit.
According to the jump method it's -2 (apparently the "animation fluff" made you misidentify it)

More specifically, taking the hitting frame as frame 0:
- The opponent can jump and attack on frame 21
- Kitana can attack on frame 22
- Kitana can jump on frame 23

From what I've checked so far, this discrepancy only seems to apply to crouching moves;
- Cyrax's d+1, d+3 and d+4 do indeed appear off by a frame.
- So far his standing moves seem correct, though I've only verified a few so far.

I'll look into it a bit more later on, but from what I've seen you just need to add a frame to the advantage for crouching moves when testing using the jump method.


p.s. The jumping input is actually to hold up for 3 frames. This doesn't make a lot of difference when buffering it after something (just make sure you buffer it early), but when jumping from a neutral state you'll see that it doesn't happen instantly. Pressing up for less than three frames causes you to not jump at all.
 
Lomyn

Lomyn

Snoctopus
My test didn't calculate difference from the attack to when that character left the ground, though, it was the difference in time between when both characters were in their first aerial frame of the jump animation. I did it this way to avoid getting confused by ambiguous looking ground movement. The difference in time between that animation frame for each character should have matched the frame advantage.
 
S

Sprint

Noob
Even the first aerial frame can be ambiguous since the animation blending lasts for longer than the jump startup. That's why I tended to measure by the landing frames, since by that point all fluffiness is out of the system.

I had a bit of work to do so I haven't looked into things much further yet. Cyrax's sweep and uppercut don't seem affected by the discrepancy though, so it may be limited to just d+1/3/4.
 
You must log in or register to reply here.