Seanson's Hitbox System v1.0.17.1 (updated june 30)

[pdk]

maybe you could consider and experiment with a system like this:

1) stats for a move (e.g. damage, IASA) could either be hard-coded or stored in a separate file depending on which way you're more comfortable with
2) instead of using another way (e.g. a 2d array as in my above example) to determine whether a move animation is available, simply test for the existence of either that move's folder or the first frame in its animation and return the "no move here" message if these aren't found
3) you could avoid even more hard-coding data by having a function that runs after the "is there an animation to show for this?" condition returns true; this function would simply take count of the frames in a move by looking for a starting frame, the frame after that, etc. for example, in a 20-frame move, it could simply look for 1.jpg, look for 2.jpg if it finds 1, etc., simply assuming that the animation is over when it can't find 21.jpg (though you probably already have another way to do this, you get the point). afterward it could return the count of frames it finds as an integer to another function (which you seem to already have in place, judging from the use of jpgs for frames), which would then construct the animation in the same way it does already (obviously with some slight changes take the previous function as an argument instead of a hard-coded value).

 

[Shmooguy]

This is absolutely amazing. If you could add shield stun frames in there you'd be making the most thorough and effective thing in smash ever.

 

[Doraki]

I wish someone would do it for the pal version, I'm sure we would see a lot of differences.
Marth's fair is slower and that's important.
Jigglypuff's dair has a lot more of lag when landing.
Some people say that falco's l-cancel lag also got increased in pal.
A lot of moves have their damage modified.
I suspect Marth's grab range got smaller in pal, it wouldn't surprise me if some hitboxes have been adjusted.
I also wouldn't be surprised if some character's speeds and falling speeds also got changed.

 

[g-regulate]

i made a post about a month ago looking for something exactly like this. thanks!!

 

[Seanson]

maybe you could consider and experiment with a system like this:

1) stats for a move (e.g. damage, IASA) could either be hard-coded or stored in a separate file depending on which way you're more comfortable with

Already done, it's hard coded.

  	if ((c==1)&&(m==0)){// Dr. MARIO JAB
  		damage1 = "3-4%, 2-3%, 5-6%";
  		recoveryspeed = "";
  		knockback = "None/Low";
  		iasa = "-";
  		notes1 = " ";
  		notes2 = " ";
  		notes3 = " ";
  		duration = 51;//LENGTH OF ATTACK (final frame number, not +1)
  	}

as you see, it is all strings, except duration is an integer.

2) instead of using another way (e.g. a 2d array as in my above example) to determine whether a move animation is available, simply test for the existence of either that move's folder or the first frame in its animation and return the "no move here" message if these aren't found

The way I have it now, will let me know which moves have been implemented, or which have been implemented + not uploaded. Another thing, is that the number of frames I have for a move isnt necessarily the length of the attack, so having duration as a variable that depicts the total length of the attack might not work. Not sure why, but I start getting memory errors after about 210 frames.

3) you could avoid even more hard-coding data by having a function that runs after the "is there an animation to show for this?" condition returns true; this function would simply take count of the frames in a move by looking for a starting frame, the frame after that, etc. for example, in a 20-frame move, it could simply look for 1.jpg, look for 2.jpg if it finds 1, etc., simply assuming that the animation is over when it can't find 21.jpg (though you probably already have another way to do this, you get the point). afterward it could return the count of frames it finds as an integer to another function (which you seem to already have in place, judging from the use of jpgs for frames), which would then construct the animation in the same way it does already (obviously with some slight changes take the previous function as an argument instead of a hard-coded value).

it seems that you are suggesting that I have the program determine the length of the attack by it finding the number of frames in the appropriate folder. I'm not seeing the benefits to doing this rather than just hard coding it in.


Doraki: the main intention of this project is to see the size/length of hitboxes. I don't know of the differences between PAL and NTSC in terms of size/length. I do know that stats are different, and it isnt really the main concern to put the stats in here (so far about 80% of the researching was done by volunteers). If there are any changes in hitboxes though, I think they should be put in. This is where volunteers come in... filmers and/or extractors for PAL, because I do not have PAL, so I would have no way of accessing that stuff. I would probably integrate PAL hitboxes (only those that differ from NTSC, probably) after I finish the NTSC hitboxes. but like I just said, we'd need someone to film.

REPORT: just taking it easy until spring break, then I can film the rest of the NTSC characters, then it is just a matter of extracting. Any volunteers that have movie editing software could be an extractor. This process will go faster with 2-3 people working on 8 chars rather than just me. I could do it, but it would take a lot more time.

 

[Zink]

Now I wish I'd gotten an AR... and a cap card... and video software... and cabling... nvm...
I wonder, has anyone found if ARs work in a Wii?

 

[petre]

yes, a GC AR works with GC games on the wii.

 

[t!MmY]

...yes, kirby (in the end) will have all variations on b attacks. this however will not be a priority as I don't think many people eat others (in tournament play). If I'm wrong correct me, I dont hang around kirby forums.

seanson, its not frequent, but certain people like those powers, namely t!mmy.

So, will all characters' moves that aren't used often in tournament play be low-priority? :-3
I've worked with video editing a lot, so I understand when you just want to have something to show rather than holding everything up. If that's the case, put in Kirby's neutral-B (Swallow animation) as a "place-holder" for the rest.

 

[Seanson]

i think it would be better to release a whole other character rather than just kirby's b variations. don't get me wrong, kirby will be finished. I think that the most viable tournament moves are the ones that aren't conditional (need specific enemy to swallow). regardless... kirby hasnt been filmed yet and when he is filmed, all his b variations will be filmed, but when they are extracted and implemented may be after all other chars are done. unless, of course, you have a few specific variations you would like to see. Dont plan on me doing especially low tier ones right away like mewtwo or bowser. i apologize if there was any confusion

just... few more.. days... until spring break. then i can film film film!

 

[blaargh198]

Finally, I've been wanting an update for some time now! And I understand if you have to wait for all of kirby's b moves, there's a LOT of them

 

[t!MmY]

Kirby isn't the only one with variable ways of using moves, so I was wondering if you're going to put in all animations of attacks? For instance, Ice Climbers when together and when separated, the change in direction with Sheik's needle storm in the air, Ness' PK Flash, Kirby's Hammer, etc.

This would, of course, require more work, so I was just wondering if you were just going to do "basic" animations, all forms of attacks, or only specific attacks with enough change to make a difference.

 

[blaargh198]

Well PK Flash already had the uncharged and charged fully hitboxes recorded. but the sheik needles don't really apply because only the direction changes, not the hitbox. When Kirby is done there will probably be both hammers because the two different attacks have two different distinct hitbox movements and hitboxes. I dunno how desynched Icy's change their hitboxes though.

 

[Seanson]

ice climbers hitboxes do not change, there is simply 2 of them. any attack you can just picture the other one doing the exact same thing.

i'm currently not working on kirby, or anything for that matter. i will probably film the rest of the chars this weekend, then throughout the next few weeks, update maybe 5-6 chars a week (if i feel productive)

all moves that have obvious changes in hitboxes are done individually. peach f-smash for example. you need to know all 3 kinds. links up+b in the air and the ground. both are used so it's important to know both. same thing for all variating attacks.

 

[cyberdemon]

Now I see why ganondorf can dair and spike you with his head...

 

[Bloshi]

For Yoshi's hitbox frames and stuff, ask Shiri. He had all of Yoshi's moves on Project Y.O.S.H.I.

 

[Seanson]

Way ahead of you

Asked shiri about a month ago, but he said that he didnt have the original footage anymore. Also, he landed during some of the aerials, which I wouldnt want to do, to keep things consistent. I'll be filming the rest of the chars sometime this week.

 

[Cort]

This is coming along quite nicely. Can't wait to see the finished product.

The interface is really boring though.

 

[Seanson]

I plan to add a knob that when you turn it, a whizzing sounds plays, and have a button that goes *HONK* when you push it.

Actually, no... hehe

But when all characters are added I will definately have some "fun" making it look pretty and more sleek. I have zero training in graphics and design. The extent of my skills is just placing everything on the window in rectangles. Maybe some graphic designers could give me a hand, if they wanted to volunteer. Let me know via PM if you want to help out the graphics.

 

[ChronoSquare]

There is only one imperfection I can find in the midst of all this hitbox capturing and whatnot... and that is that this displays the characters as if they are in a void.

When it comes down to it, the range is also how far the character would travel with the attack... for example, Jigglypuff's forward smash. The hit boxes are nothing impressive, but Puff slides a fair amount of distance across a level while using it, yet someone looking at just the hitboxes alone could not tell this...

I hear that with AR you can to the HRC and walk across the level there, and effectively measure distances. Would it be out of the question to think of recording how far character's moves travel, marking the progress of the hitbox(es) over a distance? To me, that seems like it would be more practical... especially with Jigglypuff's angular Pounding, the hitboxes tell me little that I didn't already know - what I wanted to find out was how Rising or Descending the Pound would affect the trajectory of the hitboxes... (Similarily, her grab seems so unimpressive, yet constantly I'm grabbing my friend's Doc at rediculous ranges...)

I'd start on something like that if I could, but unfortunately, I do not have the money for the items... with my friends in town, we've a way of digitalizing VCR-recorded material, but that's already between two people; we couldn't coordinate something as serious as this... (not to mention lack of any AR)

 

[Seanson]

it's true, when focusing the hitbox on the character you may not get a feel for how the player is in the environment. the hitboxes around jiggly's pound, for example, are always going to be oriented around the same place. just like every hitbox, the point of origin will never change, nor will the size (based on velocity/position/etc). I'm assuming that with the side+b, jiggly puff does not angle her fist at all, but rather just moves her body up or downwards.

the main intention of this project is for players to realize where in the character the hitboxes' points of origin are, and their size. I only need to show what can't be shown in the normal game.

 

[Tonb3rry]

What's up with the project? Stuck on Peach? =/

 

[Time/SpaceMage]

Huh? Ganondorf is listed as complete, but a lot of his moves end early (up-tilt, neutral B) or have extra white frames for about 30 more frames. In other words, they aren't showing the same number of frames as listed in the JPG directories.

I'll check the topic to see if anyone has this problem after I come back from dinner - just wanted to get a head start on finding out because this looks pretty sweet otherwise. Captain Falcon looks fine, for instance.

 

[Excel_Zero]

I downloaded it and is very nice. But how do I update it?

 

[mr_joey132]

haven't checked this thread anymore.. is falco allready in it?

 

[Nigzel]

I wish someone would do it for the pal version, I'm sure we would see a lot of differences.

I wish I had my hands on an AR. <_<

 

[FastFox]

I'll send you one from Canada, Pierre.

But you'll be paying the $2847598963295658263556.00 shipping and handling.

 

[MrHazuki]

I can't get the online version to work right now. Update?

 

[KevinM]

Yea can we get an update on this Seanson its coming along hella good.

 

[Nigzel]

THIS IS GOOD. RWB FINALLY found the AR he borrowed, WIIIIIEEEE!

Season, once I have some spare time over I'll contact you. We'll see what I can do. <_<

Joel: PHAIL, it's $2847598963295658663556.00, not $2847598963295658263556.00

 

[mr_joey132]

No it isn't... it's 2847598962295658663556.00 You obviously don't know **** about this :/

But ehm, any updates on Falco? ^^

 

[KevMoney]

Is it me or is the online verison not working?

 

[TheMegaNoob]

when ur done can you put the whole thing (with every file every move) into a file to download.
i have enough memory on my comp to hold it easily

 

[Marth/Falco Pro]

This is really neat, thanks for making it!

 

[Hank McCoy]

you forgot to add luigis taunt. please add that.

also, can you update the downloadable version, cuz i cant open the online version w/ my firefox (adblock).

 

[Hoboman725]

so, bit of a snag. I can't view the online version, God knows why. And the download page does not include the original 1.0.3.0 system...

 

[OuiCaVa11]

Can't figure out how to download this.... every download seems to be a patch, not the original version. Maybe I'm just incompetent.

 

[Stormblast]

why did this tread die?

its one of the most useful threads on the whole boards.

please continue T_T

 

[elev135]

god I have been looking for this stuff!
thanks a lot!
good luck with the rest

EDIT: STICKY!!1one1omgplx

 

[SilintNinjya]

why did this tread die?

its one of the most useful threads on the whole boards.

please continue T_T

it died because the original file doesnt exist on filefront and because the online version is broken.

 

[elev135]

it died because the original file doesnt exist on filefront and because the online version is broken.

can't Seanson just re-upload it?
I hope he does... I'm missing the first :/