Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Different curve frame sequences with hit_Fa: 14
#1
Hey guys,

I am planning to do a chase ball with hit_Fa: 14. After some testing with Julians Skull Blasts I realized that there is something really tricky to do for me. Actually, I want to have different animations depending on the t3 object flying down or up during the curve frames (50-59). First thing I thought of was a specific ik8 which is located up (or down) on the z axis of the t3 object. If the chase ball moves up to go to the curve frames it would trigger the ik8 which would lead the ball to seperate set of curve frames. But this would not work very well since the t3 object can also go up or down while flying straight. A better option could be to locate the ik8 beneath the t3 object with a matching dvx value. Since it has to slow down for the curve frames it would trigger the ik8 when entering the curve frames. But with this way I cannot determine if the t3 object is flying up or down the z axis.


Next does work in the curve frames so I basically plan to add 50-54 for curve animation (up) and 55-59 for curve animation. (down)

If possible, I need a solution where the chasing effect would be as fluent as possible and the trajectory is not influenced in any way. I also do not want to trigger the second curve animation in frame 50. I want it to happen before so that it will look smooth with the 2 sets of sprites for each curve animation.

Please note that this is just a plan. I can´t show any dc or sprites yet as I want to see first if it is even possible.

Thanks :)
Reply
Thanks given by:
#2
As far as I know, hit_Fa: 14's frame 0 is tied to 50, 1 to 51 etc, and whether it transitions from one to the other depends entirely on its x-velocity. That alone would make 50-54 and 55-59 proposition untenable, at best you'd use the 50-59 as a rotate-detecting frame and then use a separate set of frames with no homing to as the actual curving, but it wouldn't look smooth anymore unless you're also willing to use a whole tonne of frames for each individual transition.

ik8 requires touching a type 0 object to trigger, regardless ally/enemy, and would shift its position to match the centerx of the person it sticks to - essentially a teleport, so it'd influence the trajectory anyway.

Suppose you got that covered. The ball would need to create a pair of t0s with different bdys - one on top, the other below. You'd then need to have two different ik8s on the ball to check which one the ball is moving towards. That could work... until the ball goes to the top or bottom of the map; the ball on the top of the map will always curve upwards even if its supposed to curve down since the t0 above is exactly where the ball would be. And since a t0 is involved, you'd either have to slow the opointing down by opointing an object that is holding the t0 in act 1100 (invisibility) or you'd see a lot of com logos.

I considered the possibility of using ik9 instead, but it only directly sends the ball to rebound frame, and its not a solution you'd want to use for a regular projectile.

And of course, there's always the issue of 2 overlapping balls.

Verdict:
I don't think its possible to make a reliable z-axis detector to achieve the effect you're looking for, and even if it were its really not worth the effort.
[Image: uMSShyX.png]
~Spy_The_Man1993~
Steiner v3.00 (outdated), Challenge Stage v1.51
Luigi's Easier Data-Editor, A-Man's Sprite Mirrorer
Working on the LF2 Rebalance mod.
Avatar styled by: prince_freeza
Reply
Thanks given by: Arokh
#3
' Wrote:As far as I know, hit_Fa: 14's frame 0 is tied to 50, 1 to 51 etc, and whether it transitions from one to the other depends entirely on its x-velocity.

Actually I did not know that, this would make the job a bit easier. I can use frame 50-59 as some sort of start frame with an ik8 which will lead to the additional frame sequence. Next will lead to the other curve frame sequence. As far as I tested it, the curve frames do react to wait and next, as long as the x velocity is at a certain level as you mentioned. (before the ball accelerates again after the curve)


' Wrote:ik8 requires touching a type 0 object to trigger, regardless ally/enemy, and would shift its position to match the centerx of the person it sticks to - essentially a teleport, so it'd influence the trajectory anyway.

Yes, but I think it should be possible to give it a realistic look with a good matching centerx/y value. (Like the pet system which is also basically a chase ball)


' Wrote:Suppose you got that covered. The ball would need to create a pair of t0s with different bdys - one on top, the other below. You'd then need to have two different ik8s on the ball to check which one the ball is moving towards. That could work... until the ball goes to the top or bottom of the map; the ball on the top of the map will always curve upwards even if its supposed to curve down since the t0 above is exactly where the ball would be. And since a t0 is involved, you'd either have to slow the opointing down by opointing an object that is holding the t0 in act 1100 (invisibility) or you'd see a lot of com logos.


I get the issue when the ball is on the top of the z axis it would the show the wrong frame sequence then. But this can be prevented too with an other object at a higher z position which would deactivate or switch the effect of the 2 other ik8. Like the method which is used to make a ball bounce/rebound at the top and/or bottom of the z axis. The com logo is no problem anyway. I deleted it in the .exe which I use.


' Wrote:I considered the possibility of using ik9 instead, but it only directly sends the ball to rebound frame, and its not a solution you'd want to use for a regular projectile.

I considered ik9 too, but it would influence the trajectory too much in my opinion as you said.

' Wrote:And of course, there's always the issue of 2 overlapping balls.

I think this should not be a big problem too, the way I planned it to implement it into the gameplay will make it difficult to interact with another t0. The move will cost a lot of mp and only one object will be created, so an overlap with only one char is nearly impossible. the ik8 is only required during the first one or two curve frames. The only point where it could get some trouble if when 2 same characters use the move at the same time and the te objects meet exactly in the first curve frames each other but most of the time they would destroy each other normally since the object uses state 3000.


Thanks for the help so far. I hope I get some more advices from other members too. :D
Reply
Thanks given by:
#4
Quote:Thanks for the help so far. I hope I get some more advices from other members too.
I thought it would be impossible or not feasible too. There are 2 ways to force an object to go to a custom frame, an ik8 or catching, of which both of them will cause a sudden change to the object's position.

HOWEVER, since you mentioned that you fixed the "com" text appearing problem with modifying the exe, I assume you're working on a mod, and that you're okay with using custom exes. If so, I believe the dll framework has a clean solution which doesn't affect the trajectory of the energy ball (a new itr kind or something).

STILL, there's no way to incorporate an object's acceleration to how frames execute. There's no clean way to do depending on velocities either. You may circumscribe the object with itrs or something to achieve that, but it'd always break at the z-edges of the stage.


On a side note:
[Image: signature.png]
A-Engine: A new beat em up game engine inspired by LF2. Coming soon

A-Engine Dev Blog - Update #8: Timeout

Reply
Thanks given by:
#5
' Wrote:I thought it would be impossible or not feasible too.

Well I am still hoping some miraculous people would show up and present a solution which is possible with dc. :p


' Wrote:HOWEVER, since you mentioned that you fixed the "com" text appearing problem with modifying the exe, I assume you're working on a mod, and that you're okay with using custom exes. If so, I believe the dll framework has a clean solution which doesn't affect the trajectory of the energy ball (a new itr kind or something).

I am a noob when it comes to .exe editing. I just tried some things to explore a bit. The com thing and allowing the .exe to read .mp3 music instead of .wma were the only changes I made. (So that I don´t have to convert my music everytime I want to test/play lf2 by simply changing the extension of the sound files which paths are loacted in the .exe) I could never do something like this with hex.


Enough offtopic for now. If there really is no possible way with dc then I just have to deal with it.

Reply
Thanks given by:




Users browsing this thread: 1 Guest(s)