YesNoOk
avatar

Trying to apply proper hit stun and guard stun to get proper frame advantages (Read 4442 times)

Started by krudelu, January 29, 2017, 06:56:02 pm
Share this topic:
Trying to apply proper hit stun and guard stun to get proper frame advantages
New #1  January 29, 2017, 06:56:02 pm
  • **
  • (>owo)><(owo<)
Hello,

I'm trying to apply hit stun and guard stun on one of the attacks of my character and I can't seem to get the proper frame advantage.

So this is what the move is supposed to have:

4 frame startup
2 active frames
20 recovery frames

hit stun of 16 frames
guard stun of 7 frames

Given the data, I tried to apply it on the following parameters:

ground.slidetime = 16             
ground.hittime  = 16             
air.hittime = 16               
guard.ctrltime = 10
guard.slidetime = 10
guard.hittime = 10

But when I tried to test it with Training by Stupa, it gives me the wrong frame advantage value. Am I doing something wrong with the values on the parameters or maybe the "Advantage" part on the chart I'm looking at has something wrong with it? About the chart being wrong might be the case since when I did the math with their given hit and guard stun in the chart, I get the following:

-5 frame advantage on hit
-11 frame advantage on block

and this is what shows with Training by stupa

here's the link on where I'm using the frame data:
http://aquapazza.wikia.com/wiki/Frame_Data_(Touka)

5A is the one I'm doing currently
Last Edit: February 05, 2017, 08:11:56 am by krudelu
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#2  January 29, 2017, 09:45:21 pm
  • *****
  • Estoy siempre listo para un desafío.
    • Puerto Rico
    • im41784@yahoo.com
Maybe this screenshot can help you out, I got this for my
own personal notes, make sure your animation times are
Matching the start up-active-recovery times and tweak the hit
Time slide time, and pause times to get it

Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#3  January 29, 2017, 10:36:42 pm
  • ******
First off, kudos to using the actual frame data, I wish more people would take time to see how important that is.

Seems you overlooked something.

Quote
Hit Stun: Let's you know for how many frames the opponent will be stunned and prevented from any further actions, when the attack successfully connects. The varying stun values are given letter grades:

A = 12F, B = 16F, C = 18F, D = 20F.

On top of Stun there is a secondary property known as Hit Stop that occurs the same time Hit Stun does. Hit Stop are the frames in which neither player can move, these frames occur whenever an attack makes contact (hit, block, etc). Stun and Stop share the same letter grade but equate to different values, the values for Hit Stop are:

A = 7, B = 7, C = 9, D = 11F.
Hit stop and hit stun were counted at the same time, so you'd have to subtract the hit stop from the hit stun to get your actual value.

In this case you'd do 16-7, which would get you 9 frames for hit.
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#4  January 30, 2017, 09:05:47 am
  • **
  • (>owo)><(owo<)
Thanks for the input guys

I tried using this guide to do my calculation:
http://forums.shoryuken.com/discussion/106615/basic-frame-data-guide-for-newbies

So let me try to redo the math again and see the numbers...
4 frame startup
2 active frames
20 recovery frames

16 frames hit stun
10 frames guard stun

So basically from startup to the end of active frame it sums up to 5 frames since the active frames starts at the end of startup frames.

3 + 2 = 5

Then after that are the 20 recovery frames, which are after the active frames. Assuming that the move connects at the first active frame, that would be a total of 21 recovery frames.

After the move connects. the next frame is where the stun start. So basically, I subtract the respective stuns to the recovery frames remaining to figure out frame advantage.

Hit stun: 21 - 16 = 5
Block stun: 21 - 10 = 11

The leftover frames represents negative advantage. So basically

-5 frame advantage on hit
-11 frame advantage on block

And even after I made a visual to count the frames, it still end with the same result as above



And these are the values I get when I tried the move on Training by stupa. So how exactly did the chart I'm looking at ends up with -13 Advantage (assuming that the advantage the chart is referring to is the frame advantage on block)? Am I missing something?

@Niitris
About hitstop, based on it's description from the website I'm looking at, isn't it supposed to work the same like pausetime given the pausetime's function in MUGEN?

pausetime = 7, 7                 ;Time attacker pauses, time opponent shakes

I put 7 on both since according to the chart, hitstop applies on both hit and block.

Or maybe there's a particular math I'm missing that I didn't get that -13 frame advantage?

Sorry for not being clear on my first post

Also I still don't get what the .slidetime parameters exactly do so basically for now I only tried to match it with .hittime parameters in the Hitdef. I checked the Hitdef thread already and I still don't get what do they mean by "slide" in this case...
Last Edit: January 30, 2017, 12:15:56 pm by krudelu
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#5  January 31, 2017, 07:21:50 pm
  • **
    • France
Your math seem on point IMO ...

Short for "slide.time" parameters : it is used to calculate attack's pushback (on hit and on block).
Ground.slidetime is the time value (in ticks) during which the being hit character will be pushed back (if you don't define guard.slidetime, it defaults to ground.slidetime).
Ground.velocity (and "guard.velocity"), define how many pixels per tick to push back.
So your attacks pushback is (in pixels) : ground.slidetime X ground.velocity.
(actually you'll have to adjust it with the stand & crouch friction coeffs. to calculate the exact pushback, but you got the idea).


Now for the hit and block advantage stuff, I've just understood that your maths don't match what you're getting in game.
Sorry if I didn't get that info out of your previous posts, but could you please tell :
What values do you expect to have ? What values do you get in game ?


What I can say for sure, is that Mugen's (or rather KFM's) default guard states (150) give behaviours that don't match the "standard" maths. I'm even sure there is a thread about this issue but I can't find it now...

Anyway, I've been meeting a similar issue with mugen's behaviour not matching my maths.
As a side note, I've tried to sum up what I've understood about your attack with a quick visual (similar to yours :P ), with some additionnal parameters.
I've been using this kind of visuals when trying to understand what I exactly wanted to happen ingame, helped a LOT !



Now, how to get Mugen behave the way it shows in the visual above .

1) Tweak KFM's default Guard
Spoiler: KFM's 151 guard state (click to see content)

Spoiler: with small tweaks (click to see content)

As you can see, these 3 changes are rather small tweaks, but they solved my troubles with ingame results not matching my theory expectations. That's my solution and I believe there are many other ways to handle it tho.

In short, the #3  tweak prevents "losing" 1 frame during which KFM goes into 130 state before being able to punish

2) [-1]Attack's triggers and commands

Make sure your attacks can be triggered from both standing and crouching position !
What do I mean by this ?

Spoiler: KFM's St. Light Punch (click to see content)

This attack cannot be activated directly from a crouching position, even if not holding down anymore.
The engine will "LOSE" 1 frame going through standing state (0) before launching the attack !

Spoiler: adjusted St. Light Punch (click to see content)

With the commented adjustment, the standing LP will be able to be performed from a crouching position without losing that 1 frame.

And last, I personnally use a "larger" default buffer time for buttons in my cmd file : 3 instead of 1 for command.buffer.time.
Code:
[Defaults]
; Default value for the "time" parameter of a Command. Minimum 1.
command.time = 15

; Default value for the "buffer.time" parameter of a Command. Minimum 1,
; maximum 30.
command.buffer.time = 3

It helps for reversal attacks, even if it's a normal move, and lets you perform "frame perfect" punishes when getting out of blockstun.

I hope all this helps a bit.
Sorry for the length  :P
- samurais, swords & stuff ? → Another Blade
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#6  February 02, 2017, 02:29:06 am
  • ******
About hitstop, based on it's description from the website I'm looking at, isn't it supposed to work the same like pausetime given the pausetime's function in MUGEN?

pausetime = 7, 7                 ;Time attacker pauses, time opponent shakes

I put 7 on both since according to the chart, hitstop applies on both hit and block.

Yeah, that's exactly it.
Those 7 frames would be the pausetime, and the remaining 9 frames would be ground.hittime. That's how you'd get 16. And your advantage should come out to -13 (or -12 depending on other factors).

Guard seems fine. I assume that hit stop isn't a part of guard stun, so the 10 number should be good.
Last Edit: February 02, 2017, 02:37:45 am by Niitris
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#7  February 02, 2017, 11:09:42 am
  • **
  • (>owo)><(owo<)
Yeah, that's exactly it.
Those 7 frames would be the pausetime, and the remaining 9 frames would be ground.hittime. That's how you'd get 16. And your advantage should come out to -13 (or -12 depending on other factors).

Guard seems fine. I assume that hit stop isn't a part of guard stun, so the 10 number should be good.

I suppose if what the chart meant by frame advantage is frame advantage on hit. Tried this for 5A but when I get to others, I tried what you mentioned and it gives me numbers that are off from the frame advantage values that the chart mentioned

If what the chart on the site meant by frame advantage is on hit, then it seems that the case is that frame advantage on block is more advantageous than frame advantage on hit, which doesn't makes sense to me since recovering on hit should be harder than recovering on block. Let me try applying it on 5C:

Since the hitstun for that is 20 with hitstop of 11, it will be:
pausetime = 11
ground.hittime = 9

And by the time I try it with training by stupa, I get -31 frame advantage, which is really off from the supposed -23 frame advantage

When I looked at most (if not all) fighting games frame data, when they include frame advantage, they state both frame advantage on hit and on block. But whenever they don't state either of those, it's usually considered "on block". Even games that have combos like Guilty Gear and Blazblue does this.

Even Arcana Heart games, which is one of Examu's games which the same company that made Aquapazza, have frame advantage on hit more advantageous than frame advantage on block.

I guess I may have to just take certain parameters into account in one of the two ways
A. Startup, Active, Recovery, Hit Stun, Guard Stun (This seems like a better option)
B. Startup, Active, Recovery, Frame advantage (considering on block), Hit Stun

When I applied Option A above, considering the frame advantage the chart mention considering it's frame advantage "on block", the values I'm getting is just off by 1 except for 5A, which is just off by 2

I guess hit stop would most likely function as pausetime just like what mr_bourrepalestick mentioned above in the picture he made

I guess one way to remedy this is to add more recovery frames or reduce the guard.hittime than usual to be able to match the frame advantage according to the chart, but a difference in 1 frame (or 2 frames) can decide the outcome.

Or maybe I should give the character I'm making 1 frame advantage ahead compared to the chart, which is what I currently have based on the math I did considering Option A above just for the sake of getting the character to MUGEN despite it being custom (consider it a "buff" for MUGEN on my end I suppose)?


Any thoughts on this?



Last Edit: February 02, 2017, 12:24:59 pm by krudelu
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#8  February 02, 2017, 11:20:00 am
  • **
  • (>owo)><(owo<)
@mr_bourrepalestick Thanks for telling me about the ground parameters. About the math I did, I've been getting those values I got when I did the math  just like how I expected them to be when testing it on Training by stupa ingame.
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#9  February 02, 2017, 12:32:37 pm
  • ******
    • www.justnopoint.com/
If MUGEN is being weird you can always go the @felineki: route and make every attack that lands a custom state to be sure the opponent moves correctly each tick. That's dedication!

Otherwise you just need to mess with the number to get the frame advantage you desire because MUGEN deals with hits different than most fighters.
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#10  February 02, 2017, 06:04:03 pm
  • **
  • (>owo)><(owo<)
I'll take note of this when I end up bumping to it in the future. As of now, I just kind of starting to find the frame data chart I was looking at to be kind of inconsistent with the givens provided but when the math is applied, I get the numbers I expected and those numbers show up in MUGEN so so far so good when I did the math myself when going with Option A when doing how frame data is typically calculated.

I'll leave this topic opened for now until then

Thanks for the input guys
Last Edit: February 02, 2017, 06:18:47 pm by krudelu
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#11  February 02, 2017, 06:29:23 pm
  • ******
The chart didn't do a good job explaining things, but yeah, it was frame advantage on block. Those kind of charts normally have both if it's necessary, which it would've been here. 5A just happened to add up to -13 with the way I did it. :P
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#12  February 02, 2017, 06:44:13 pm
  • **
  • (>owo)><(owo<)
I suppose. As for now, I'll go with Option A then
Re: Trying to apply proper hit stun and guard stun to get proper frame advantages
#13  February 03, 2017, 12:39:23 pm
  • **
    • France
@mr_bourrepalestick Thanks for telling me about the ground parameters. About the math I did, I've been getting those values I got when I did the math  just like how I expected them to be when testing it on Training by stupa ingame.

Glad to hear you've found your way out of your issue.
Good luck mate :)
- samurais, swords & stuff ? → Another Blade