The Akira Thread

[Goro Hazuki]

Mini update: I'm currently going through a full break-down of Akira in DOA5. I will update the first page with his strengths, weaknesses, and full move list breakdown (frames, crushes, tracking, safety) over the next two days. Please let me know if there is a specific point you are interested in me covering or if you have something specific/fun/cool you want to add.
I already assembled a fun list of things with him, so I hope you guys find it informative and helpful.

 

[Goro Hazuki]

First Post updated. (More to come soon). Let me know if you guys are happy with the direction I'm taking and share what topics you want covered. Input and ideas are greatly appreciated, so let's all level up our game with Akira together.

 

[Ghosty-J]

I'm loving the update you did for the first post. I could have sworn that was 15 frames though, I could be wrong though. This will definitely help me turn my losing streaks around. Just need more time to practice though XD

 

[Goro Hazuki]

I'm loving the update you did for the first post. I could have sworn that was 15 frames though, I could be wrong though. This will definitely help me turn my losing streaks around. Just need more time to practice though XD

Thanks for the feedback. I hope it's useful.
Do you mean 15 frames for execution AND hit, or just execution?
I just double checked the game is indeed 14 frames execution. Add one more for hit, and you have 15. So it depends on what you're measuring. I am sticking with what the game provides (the execution) in the thread because it is the most accessible to anyone reading/playing.

 

[Ghosty-J]

Yeah, I just misread or my memory fudged up on me. Thanks for the heads up lol

 

[Extra]

Nice guide.

For (backturn), is guaranteed followup I think. Even on fastest the AI can't escape out of it. The timing is pretty strict, but because of backturn, launches for , combo.

 

[Goro Hazuki]

Nice guide.

For (backturn), is guaranteed followup I think. Even on fastest the AI can't escape out of it. The timing is pretty strict, but because of backturn, launches for , combo.

Thanks for the input Extra. Unfortunately, that's not guaranteed. The back turn is a maximum of 15 frames in fastest recovery. The uses 14 frames execution +1 for hit. So it's just shy on speed. Also, the distance of the back turn creates another challenge.

 

[Extra]

Are you sure? With the AI on fastest stagger escape and guard on reaction, it cannot block the elbow. Also, you take enough of a step forward to cover the distance.
In theory, if it gives you 15 frames advantage, I don't *think* you can block on frame 15 - thus a 14+1 move should hit.
Give it a try.

Edit: Here's my crappy video showing it in action.

 

[Goro Hazuki]

You set the dummy on fastest recovery, but the recovery is still on the off position

 

[Extra]

Ah you're right. I went back to training and set it to on.

Here's another video:

 

[Goro Hazuki]

Yeah, that's exactly what I've been getting. If you notice, the elbow does not combo. It combos in at fast recovery, but not fastest. From my experience, good players stagger escape in between fast and fastest, so landing it in a real match is tricky. But I do think it is our best option at the moment. Of course, it doesn't hurt to set the opponent a few times with the elbow, then mix them up with a for high counter throw damage when they hold in anticipation.

Thanks for the effort with the videos though!

Edit: On fast recovery, you can do this combo:
,,(),, (98 Damage)

I'm hesitant to post this in the combo thread because it is misleading and kinda easy to escape the second hit (even if it's on fast recovery).

 

[Extra]

I'm having the suspicion that the elbow is actually landing when the opponent is "turning around", which is why it doesn't combo. But I'm still not convinced that they can escape/step/block it.

And yea, landing it in a real match would be extremely tricky.

Edit: So anyway, theorycrafting time (elaborating a bit more on my thoughts).
Let's assume that at fastest stagger escape, the opponent is freed before lands. According to your frame data suggestion, they might have a 1 frame window where they can do something.
It's already shown that they cannot guard it. My theory is that you cannot guard from a backward position instantaneously, so they'll get hit in the process of.
So in order to defend against the elbow, they need to somehow evade it. I don't think 1 frame is enough time to start dashing away from the range of the elbow. I know sidestep avoids attacks from a front position starting from the very first frame, but what about back position? (Brad Wong? Does he have a back-facing sidestep?) I imagine it would be pretty silly if people can sidestep from a back position instantly from a gameplay perspective, so my suspicion is no. But I will have to test that when I get home from work later.

Now my proposition is this: if there's no defense to an attack, then it is guaranteed, regardless of whether the system registers it as a "combo".

 

[Goro Hazuki]

The game registers combos on BT opponents (even during turn around stun). Check Kasumi's for reference. In Akira's case, your opponent may very well be likely to stagger escape and guard in time. They may not be able to hold in time (needs testing with a friend), but I'm quite confident that they can turn and guard in time. I don't have guard frame data, but I believe it is instant.

All that being said, given an opportunity in a match, I would use the 666P elbow to try an sneak in the hit. Benefit of hitting from the back (even not under stun) is that the float properties are different to hitting from the front, and you are able to juggle after the elbow.

I do like your proposition, but I would hate to get smacked in a competitive environment by a Bayman/Bass/Leifang/Tina throw because they guarded in time.

 

[Extra]

I do like your proposition, but I would hate to get smacked in a competitive environment by a Bayman/Bass/Leifang/Tina throw because they guarded in time.

I don't think you''ll have too much to worry about. If the CPU can't slow escape and guard in time on fastest setting, I'm not sure any human can.

 

[wali]

damn i just tried the game with friend ,it looks like i well need more time than i thought to seprate DOA Akira from his VF counter part .

 

[Goro Hazuki]

I don't think you''ll have too much to worry about. If the CPU can't slow escape and guard in time on fastest setting, I'm not sure any human can.

Ok, I tried a few things out. The definitely does not connect as a combo after counter hit . But I did find a guaranteed follow up. It's slightly similar...

You can get a guaranteed follow up when doing the move on counter from an open stance which causes less travel in your opponent. In this very specific situation, Akira can follow in with a very quick ( is 1 frame too slow). Unfortunately, no further guaranteed follow up to has been found at the moment.

*added this to front page*
Adding section on throws tomorrow. Is there something specific you guys want added to the front page before other sections (i.e. ground game)?

 

[Extra]

Interesting, I'm gonna have to test that some.

I did some testing myself, on the theory of backturn guard speed, and found some interesting things.

The usual AI settings...COM reaction = Guard All, stagger escape = fastest.

First things first...instant double palm (13+1 frames) will connect as a combo if you buffer perfectly. This will work whether in open or closed stance.
However, even if you don't buffer it perfectly (i.e. it doesn't register in the combo counter), there's a small window where instant double palm will still connect. This reaffirms my theory that you cannot backturn-guard instantly.

But why stop there? I need to know exactly how big this window is.
So through some testing, I found the slowest move I can connect (that doesn't count as a "combo") is another , which is 18+1 frames (other things that also connect: , , ) . That is SLOW. However any 19+1 frame moves (such as or ) will not connect period.

So, from this, my preliminary conclusion is that it takes 6 frames for the AI to turn around and guard (assuming AI is "free" to act on frame 15). It takes even longer for the AI to turn around and counterhold. This actually seems to be in line with what I expect. If I recall correctly, it also takes exactly 6 frames for a character to turn around and guard in VF5FS.

I need some help to verify this though. Currently I don't have anyone to play this thing with offline, so if any of you guys have a friend that you can test this with - see if they can escape any move faster than 18+1 frames. In the main time, I'll look for some other DOA character with a similar back turn attack and see if I get similar results from them.

Edit:
Just talked to Mr. Wah to see if he had any experiences with backturn speed, and he said this:
Bass' BT ground pick up throw is +11 frames. After picking the opponent off the ground, his (16+1 frames) will not connect, which is 7 frames later, assuming the opponent is free to act on frame 11. However, a is guaranteed, which is 14+1 frames, that puts it at 5 frames after the opponent can act. This seems to correspond with my findings. I'll keep an eye out on further data.

 

[Goro Hazuki]

Good stuff Extra. This makes me hopeful. I will do some testing as well (with a friend) to see what we can connect, this would be a nice opening to Akira's back turn stun game if possible.

 

[Goro Hazuki]

Double post: Extra you are the man! I set the AI on record to do ,repeatedly at two different speeds. One at fastest input and one with a very slight hesitation (replicating a real life match situation with a slight delay of "ok it landed, now to follow up".
On fastest input, while the combo would not register, the elbow hit every time
With the very brief hesitation (and I'm saying very brief), I escaped and blocked/evaded every time.
Now granted I am expecting the attack so I am getting fastest recovery every time. It's nice to know that this connects. It just puzzles me that the game doesn't put it as a combo, so I always felt uncertain about it. But I guess your theory about this seems to have been right.
Good brain crunching! This is what this thread needs, more collective ideas and help from members.
Let me know what you find with your exploration as well before I update the front page. Cheers.

 

[Extra]

Thanks
I also feel this is what forums are for, to discuss and find out stuff that we might not individually. Everyone benefits!