DOA5α Hold Frame Data

[grap3fruitman]

Here's a list of the frame data for each character in the DOA5α demo. This is ripped out of the game's code and it doesn't exactly tell you which hold(s) the data belongs to, so figuring that out is left to manual testing.

Name	Loc	Initial	Active	Recovery
Ayane	148816	0	20	10
Ayane	148832	0	22	4
Ayane	148848	0	18	17
Ayane	148864	0	20	10
Ayane	148880	0	18	17
Ayane	148896	0	20	10
Ayane	148912	0	18	16
Ayane	148928	0	20	6

Name	Loc	Initial	Active	Recovery
Ryu	140960	0	20	10
Ryu	140976	0	22	4
Ryu	140992	0	18	17
Ryu	141008	0	20	10
Ryu	141024	0	18	17
Ryu	141040	0	18	17
Ryu	141056	0	18	16
Ryu	141072	0	20	6

Name	Loc	Initial	Active	Recovery
Hayate	120816	0	20	10
Hayate	120832	0	22	4
Hayate	120848	0	18	17
Hayate	120864	0	20	10
Hayate	120880	0	18	17
Hayate	120896	0	18	17
Hayate	120912	0	18	17
Hayate	120928	0	20	6

Name	Loc	Initial	Active	Recovery
Hitomi	112400	0	18	12
Hitomi	112416	0	18	12
Hitomi	112432	0	18	17
Hitomi	112448	0	20	10
Hitomi	112464	0	18	17
Hitomi	112480	0	20	10
Hitomi	112496	0	18	16
Hitomi	112512	0	20	7

Two points:

-According to the Ripten interview here, expert holds have more active frames and less recovery frames than regular holds. This leads me to believe that the 0/20/10 holds are expert holds.

-According to Berzerk's testing here, normal holds appear to be 0/18/17, which coincides with the data pulled from the game. All the characters but Hayate seem to have a hold that's 0/18/16, I'm speculating that this is the normal low hold.

 

[UncleKitchener]

I'm guessing eight is the max number of holds characters can have.

 

[grap3fruitman]

It should be more than that, I would think. I need to double check into these numbers and make sure I'm not missing any. The data isn't exactly made for human consumption and kind of difficult to parse. I'll report my findings later tonight.

 

[MASTER]

Interesting but shouldnt all characters only have 5 holds? Only exception is hayabusa with the advance holds i would think but that would only make 7 holds total right? how is there eight?

7F High
4F Mid
1F Low
46F Adv, Mid,K
64F Adv, Mid, P

Im wondering about what shimbori said recently saying they did a lot of testing with different hold mechanics in the game before coming down to the 3 point hold. Is it possible that those different kinds of holds are in the code and not the actual game??

Very interesting stuff either way. I will need to do more testing but i thought i had narrowed it down to

Normal hold = 0/18/17
Advance hold =0/18/12

*In stun*
Normal hold = 0/16/19
Advance Hold = 0/16/14

Back to the lab >_<

MASTER

 

[grap3fruitman]

A high punch hold is counted separately from a high kick hold. At least that's how it was with the older games. It should be the same here but like I said, I'm looking into this more today. I haven't had a chance yet.

 

[MASTER]

New findings. It seems to be true that some characters have different lengths of holds. Right now i cannot figure out what the window of the hold is or the recovery since i need a partner for that but i will hopefully get to fully test that tomorrow. However i can at least find out the total length of the animation.

Hitomi's normal holds out of stun are:
High hold = 35 in stun its still 35 (active 16?) Seem to get that for active, need more testing.
Mid hold = 35 in stun its still 35 (active 16?)
Low hold = 34 in stun its still 34 (active 16?)

Hayate
High hold = 35
Mid hold = 35
Low hold = 34

Hayabusa
High hold = 35
Mid hold = 35
Low hold = 34

I will try to update my post or quote it later if i find anything new.

MASTER

 

[Matt Ponton]

are you getting your data from trial & error in-game? or just recording video data of it and reading it frame by frame?

 

[grap3fruitman]

So the data in the OP looks accurate. There could be more hold data elsewhere in the game's code but this is all that was found immediately. It's not exactly easy to sort through.

 

[MASTER]

are you getting your data from trial & error in-game? or just recording video data of it and reading it frame by frame?

Currently trial and error is the best way for me.

@Grape, The OP has some truth to it and it seems that there are some that can be different but i havent ran into anything that adds up in particular with some of those numbers with the exceptions of the ones I previously mentioned before this thread. More testing is needed of course =/.

MASTER

 

[DrDogg]

18-22 frame active window is way too large given the frequency with which you can use counters.

LOL at 4-6 frames of recovery on some of those.

 

[synce]

I've never been big on counting frames but the improvement in DOA5 is noticeable even to the naked eye.

In DOA4 I've played guys that threw out 3 counters before my next move came out lol

 

[Allan Paris]

damn, doa5 looks so much better.

 

[x Sypher x]

I've never been big on counting frames but the improvement in DOA5 is noticeable even to the naked eye.

In DOA4 I've played guys that threw out 3 counters before my next move came out lol

That's only a minor improvement, it's still kinda spammable and needs a few more frames of recovery and few less in the active. Interesting video though.

 

[MASTER]

That's only a minor improvement, it's still kinda spammable and needs a few more frames of recovery and few less in the active. Interesting video though.

Depending if we ever get the actual frame data from TN there is a chance they wont need to do nothing else but add the 4th point hold. Only spamable right now because there is no difference in mid attacks.

Plus the punishment for spamming in this game outweighs the hold already. The Data that Grape was able to post up still seems sketchy and i haven't found anyone that has a 4 or 6 frame recovery unless they wanted to count hitomi's or ryu's parry into the equation but even then the recovery still seems longer than that.

MASTER

 

[Raansu]

I've never been big on counting frames but the improvement in DOA5 is noticeable even to the naked eye.

In DOA4 I've played guys that threw out 3 counters before my next move came out lol

Most of us that played dimensions and the doa5 demos noticed the changes right away, but it's still not enough.

 

[synce]

Well they could tighten it up a tiny bit more but if they left it like this I wouldn't mind. In the end it'll always be a guessing game, I know some people can't accept that but that's why there's VF

 

[x Sypher x]

Well they could tighten it up a tiny bit more but if they left it like this I wouldn't mind. In the end it'll always be a guessing game, I know some people can't accept that but that's why there's VF

While it is true that DOA is the only series considered to be a true guessing game because of it's universal hold system AND because it's the only fighting game that allows you to escape a stun, which sets it apart from all others, that doesn't mean it can't make a change for the better. If this problem was fixed, sure it would become more akin to VF in a sense, but it will still be DOA. The guessing will be severely toned down and restricted to more educated guesses, which other fighters properly culminate, and will become more viable in the competitive scene. That along with other minor concerns.

 

[Matt Ponton]

Only spamable right now because there is no difference in mid attacks.

I can assure you sir that if you are admitting that it's actually spamable right now, the difference of me spamming , , , and , , , is not what allows me to 'spam' it. The only thing that is affected is the chances of randomly connecting a hold, however that still doesn't affect the conditions on when I can hold after performing another hold.

 

[DrDogg]

Depending if we ever get the actual frame data from TN there is a chance they wont need to do nothing else but add the 4th point hold. Only spamable right now because there is no difference in mid attacks.

Plus the punishment for spamming in this game outweighs the hold already. The Data that Grape was able to post up still seems sketchy and i haven't found anyone that has a 4 or 6 frame recovery unless they wanted to count hitomi's or ryu's parry into the equation but even then the recovery still seems longer than that.

Enjoy your 15-man tournament scene.

 

[grap3fruitman]

Tournies with 15 people are easier to rig.