Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!
You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!
It appears that you are using ad block :'(
Hey, we get it. However this website is run by and for the community... and it needs ads in order to keep running.
Please disable your adblock on Smashboards, or go premium to hide all advertisements and this notice. Alternatively, this ad may have just failed to load. Woops!
If you're going to do that, I think it'd be better if you could just make it parse http://opensa.dantarion.com/wiki/Article_Floating_Points for the offsets, values, and descriptions on launch. Then, if someone updates the float page, the definition of it will be updated.
If you're going to do that, I think it'd be better if you could just make it parse http://opensa.dantarion.com/wiki/Article_Floating_Points for the offsets, values, and descriptions on launch. Then, if someone updates the float page, the definition of it will be updated.
bug in PAT0 parser.
Open up FitDededeMotionEtc, Scroll down to the PAT0 stuff, expand the first node, watch as BrawlBox freezes trying to populate the list.
The bug is in the "CleanTextures" function...it hink...
Anyways, as a result, I cannot call FindChildrenByType on that .pac, possibly others as well.
bug in PAT0 parser.
Open up FitDededeMotionEtc, Scroll down to the PAT0 stuff, expand the first node, watch as BrawlBox freezes trying to populate the list.
The bug is in the "CleanTextures" function...it hink...
Anyways, as a result, I cannot call FindChildrenByType on that .pac, possibly others as well.
I've got a question. When you pop out the right tab, there's a button saying "load external moveset file". Which file is the external moveset file? I tried FitCharacterMotionEtc.pac, which gave me an error, and FitCharacter.pac which didn't give me an error, but nothing showed up except for some hurtboxes on the third tab. So which file is the actual "external moveset file"?
Will the next BrawlBox version fix up PAT0 editing? The older version has a bug, in which you can't properly edit PAT0 files; it's likely old news though.
There was only hurtbox editing v0.66. And you couldn't even add new hurtboxes or delete any.
v0.67 can fully edit scripts, parameters, articles and misc data and save everything to a rebuilt file.
Will the next BrawlBox version fix up PAT0 editing? The older version has a bug, in which you can't properly edit PAT0 files; it's likely old news though.
I hate to bring bad news, but v0.67 still has the PAT0 editing bug.
Edit: After some extra tests, it appears that you still can't properly modify PAT0 files while using the Advanced Model Editor. You have to create the entries manually, using the New Entry command, but at least the entries no longer appear as <null> when you save the PAT0 file.
Edit 2: Okay, I came across this error while testing out the PAT0 editing...
Code:
************** Exception Text **************
System.NullReferenceException: Object reference not set to an instance of an object.
at System.Windows.Forms.PAT0Editor.TexChanged(Object sender, EventArgs e)
at System.Windows.Forms.StringInputComboBox.Apply()
at System.Windows.Forms.StringInputComboBox.OnTextChanged(EventArgs e)
at System.Windows.Forms.Control.set_Text(String value)
at System.Windows.Forms.ComboBox.set_Text(String value)
at System.Windows.Forms.ComboBox.UpdateText()
at System.Windows.Forms.ComboBox.WmReflectCommand(Message& m)
at System.Windows.Forms.ComboBox.WndProc(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.OnMessage(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.WndProc(Message& m)
at System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)
I also tried PAT0 editing with eye lids that require palettes.
I hate to bring bad news, but v0.67 still has the PAT0 editing bug.
Edit: After some extra tests, it appears that you still can't properly modify PAT0 files while using the Advanced Model Editor. You have to create the entries manually, using the New Entry command, but at least the entries no longer appear as <null> when you save the PAT0 file.
Edit 2: Okay, I came across this error while testing out the PAT0 editing...
Code:
************** Exception Text **************
System.NullReferenceException: Object reference not set to an instance of an object.
at System.Windows.Forms.PAT0Editor.TexChanged(Object sender, EventArgs e)
at System.Windows.Forms.StringInputComboBox.Apply()
at System.Windows.Forms.StringInputComboBox.OnTextChanged(EventArgs e)
at System.Windows.Forms.Control.set_Text(String value)
at System.Windows.Forms.ComboBox.set_Text(String value)
at System.Windows.Forms.ComboBox.UpdateText()
at System.Windows.Forms.ComboBox.WmReflectCommand(Message& m)
at System.Windows.Forms.ComboBox.WndProc(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.OnMessage(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.WndProc(Message& m)
at System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)
I also tried PAT0 editing with eye lids that require palettes.
Ok, thanks for telling me. I didn't modify any code in the editor itself because I thought the errors were only in the rebuilder. Looks like I gotta revise the editor now too lol
Another bug (not sure if it's already known):
Event 00070100 ("Subroutine") causes the Moveset Panel to incorrectly read the first Timer in a move's event list. It performs the actions after the first timer 1 frame later.
So for instance, P:M Pit's Fair (which has Event 00070100 listed first) is supposed to come out frame 9. BrawlBox displays the hitboxes frame 10.
Another bug (not sure if it's already known):
Event 00070100 ("Subroutine") causes the Moveset Panel to incorrectly read the first Timer in a move's event list. It performs the actions after the first timer 1 frame later.
So for instance, P:M Pit's Fair (which has Event 00070100 listed first) is supposed to come out frame 9. BrawlBox displays the hitboxes frame 10.
Alright, thanks for telling me.
Subroutine and Goto events are still in a bit of an experimental stage though.
I'm going to revise how they work in a later release.
ok...
I've been working on this post offline, so it's REALLY large >_>
but it has quite a few interesting tidbits of info as well as some ideas and implementations for brbx
BJ ... your brbx is extremely useful =D
Yes there's no mistaking it, that's Pachi, and he's fully working.
to see the horors I went through, open the collapse >_<
I just said screw it and decided to work with the model Roo gave me
^once I saw that, I did this:
- cried for 5 minutes
- flipped the table so hard it never landed (during a large rant)
- sat back down at my compy to realize I was a blooming idiot >_<
(pathetic... I know)
it's impossible for the model to work properly with less than the amount of the original bones...
@Everyone who told me this: you were right dangit >.<
the model roo gave me still had it's own problems...
(you had to have at least 1 of Pika's bones for it to work at all, of course)
- you couldn't hold your grab (but you could throw) (no ThrowN)
- you could be grabbed, but being thrown splattered the polygons (no HipN??)
(you could at least get out of the splatter to foot-stool the opponent... heh)
- you could pick up crates/barrels but not capsules and some other items >_> (no HaveN)
(you could hold the beamsword) =D
- you couldn't hold to run/walk, and you couldn't roll (didn't have TransN to YRotN)
I figured the bone indexing thing out after this... heh
everything works perfectly now
also...
I found a small (large) bug that required me to re-adjust the UV attributes
luckilly I had my HexEdit template or I'd've said screw it... D:
when converting from v9 to v11, the UV attributes are reset,
sadly, Brbx doesn't allow manual data adjustment... >_<
SO... I tried replacing the objects to adjust it... BAD MOVE!
doing that seems to remove the object's texture completely, even though the material is assigned to it...
another attempt ended up completely removing the bone transformations... but I figured that one out...
this is what the original attributes were
after hexing them back, I got this:
^I've fixed the eyes since this image.
I'm not worried about it too much as this thing's nowhere near release yet
(actually all it needs is animation... heh)
^in other words... it works perfectly
although I can't get the PAT0 animations to work after redoing his eye textures in Brawl's format >_>
and another (small) bug when adding keyframes to the CHR0 files alone...
heh... I was trying to add up the translations up to 'origin' and then inverse that to inverse transform the model...
but that didn't quite work >_<
(neither did my calculator which reapplied the initial transform before subtracting)
^best results with that though, but the model still slightly jerked around D:
I NEED THE BONE BACK AT IT'S T-POSE POSITION!!!
WHY IS THAT SO HARD TO ASK FOR!!! :srs:
other than that un-interesting tidbit of info...
I have a few requests that might actually make Brbx into something USEFUL for animation: :srs:
(actually there's quite alot of requests) >_>
NOTE: these requests go to anyone who can program them in.
Previewer window:
- update all (shown) models that use the animation
I only want to re-animate 1 model, but I want 1 other model shown (out of 5)
(I don't want to have to show every single model to re-animate just 1)
- (personal) highlight the active bones for the keyframe
very useful for animation cloning...
- edit a single bone for every keyframe (globally)
(eg: 'Origin' should be L,R,S=[0,0,0],[0,0,0],[3,3,3] (from the global origin) for frame 1 of every animation)
yes I'm making this request AGAIN... -.-*
^
- keyframe shifting
(if a KF isn't exactly in the right spot)
- snap to selection, grid, or center
(would make brbx almost as useful as Blender)
NOTE: this goes for anything in the 3D grid space (bones, objects, verts, etc...)
- invert Keyframe data
(animate one step, then copy and invert the data for the other)
idea from Blender
- front/side/top view keys (ctrl+key for inverse view)
idea from Blender (numpad: 1,3,7)
- perspective/ortho toggle
perspective is much more annoying to animate in than ortho
(and it's not like it's hard to implement either) :srs:
idea from Blender (numpad: 5)
- use Mouse Wheel to change selected LRS entry values
- (personal) display grid instead of floor
display modes:
X/Z 10x10 standard grid when viewing normally (rotate the model around)
X/Z full grid for top view
X/Y full grid for front view
Z/Y full grid for side view
(make the X/Y/Z origin lines either bold or colored plz)
Main Window:
- instead of having to open a new window to rename an entry (referred to as a "Node" in brbx)
why not create an entry widget in place of the displayed text based on a click1 - click2 (not dbl-click) event.
(just like in windows explorer)
- advanced user mode (for people who know what they're doing (like me, lol))
(allow editing data such as CP Register flags)
- (advanced mode) a hex editor right-click option (for any data section)
- ^should possibly implement a built-in template for the selected section as well (so you're sure of what you're doing)
- (advanced mode) allow unhandled (anonymous/unused) data
for when you want to add verts and other sub-object datas before the object data itself
(weather it be through the hex editor or the main GUI)
TIP: try to avoid majorly repeating a process (click this -> click that -> click OK -> Repeat)
that just creates a major annoyance...
as well as something I personally don't have the patience for -.-*
also...
can do me a favor BJ...
please make sure your terms actually sound like you know what your doing >_<
- PSA/Moveset == Game Logic
- Shader == Material Node (I've compaired your new finding's with Blender... again, and now am 100% sure about this)
- Display List == Shader (the "Display List" is everything OpenGL draws during it's loop)
btw...
I was wrong about shaders... <:/
the fur I was thinking of is done by partical systems which the wii doesn't seem to support >_>
the shaders themselves just re-adjust light values and apply textures in a similar way to material nodes...
^(just not as advanced)
material nodes play with the shader's returned values to give it a slight adjustment based on the view or other inputs
^(similar to a shader rendering "glass" (I think) >.>)
also... 1Q...
samus's suite reflection texture...
would that be handled through the shader or the node?? (just curious)
... I think it's the XF registers in the shader (partly) that do it...
(knowledge from Pachi's specular texture and Pikachu's metal materials)
and now for another thing...
by chance, would you be interested in adding my (highly experimental) dynamic mesh modelling system into Brbx??
if so, be warned, this may greatly reduce rendering performance due to the required quality of the models (when used)
you should set a resolution adjustment var to decrease the number of rendered verts
^does not affect overall model shape
this shouldn't interfere with basic model rendition
what is my modelling system??
well, I must first say the idea behind it was based off blender's sub-surf modifier and curve-surfaces...
the only problem with curves though was the gridding system used.
the curves require a U,V ordered vertex array:
Code:
U_array = [
[ [XYZ],[XYZ],[XYZ] ],
[ [XYZ],[XYZ],[XYZ] ],
[ [XYZ],[XYZ],[XYZ] ]
]
for V_array in U_array: #rows
for Verts in V_array: #columns
thus you could initially draw something like this:
^I've only slightly figured surfaces out... heh
I can't fully support them in UMC yet >_>
what the system doesn't allow you do is:
U_array = [
[ [XYZ],[XYZ],[XYZ] ],
[ [XYZ],[XYZ],[XYZ] ],
[ [XYZ],[XYZ] ]
]
the error: TypeError: incompatible types, c_float_Array_3_Array_2 instance instead of c_float_Array_3_Array_3 instance
basically saying the grid doesn't support triangles...
my system attempts to fix this by using an array of verts, lines, and primitives with a few extra operations such as:
- line strength, specifies an amount to pull the outer curve by (the reason for the array of lines)
- vert resolution, which specifies how many sub-verts are allowed in the inner curve
- vert strength, which specifies how much to pull the inner curve by
there are still more operations to be found...
(I can't think of everything verts are used for)
a simple square rendered as a circle:
a triangle rendered as somewhat of a circle:
a simple, yet complex shape:
basically, the inner curves are owned by the verts while the outer curves are owned by the lines
the outer curves get their vert counts from a divized sum of the resolutions from the verts belonging to the curve's lines.
in other words:
res1 = 5
res2 = 7
outer_res = int( (res1+res2)/2 )
Blender's subsurf modifier wasn't all that great either as it used a precalculated sub division to round it's surfaces.
(can't achieve an actual circle or sphere)
another advantage my system has over this is how you can specify an amount of
how much influence the vert or line has toward the sub-mesh. (as discussed in the vert/line influences above)
I do have an idea I'm thinking about scrapping:
^uses a bone to create a skin affect between 2 dynamic models.
I'm thinking about scrapping it because it's just too raw >_>
you may end up linking more than what you want through a transformation
say you're doing this for an arm...
if the hand was to come w/in range of the link, then the whole arm would link
using it on the mesh instead of the bone may result in the same outcome, if not worse...
(the whole model looking as if it were in a balloon w/o air)
but those may only minor errors that could possibly be fixed using a few other options...
so I'm unsure about keeping it >_>
as for what I have already...
here's a few tests I've done while building this (the most recent at the bottom)
(rmbr I can't support an actual solid mesh with curves)
^just a basic point to curve plot
^the curves are drawn by the vertex order...
^this uses face centers for start/end points and the common line center for the midpoint
^same model with the top face added
^2 linked faces (one shorter) flat against each other (testing for invalid curve, but got perfection)
^went back to my other model to support line influence (0.6 to 1.0) and vert resolution
^supported line influence (0.0 to 0.4)
^added lines to where the end-outer curve should be if there's no linked faces
^after a bit of restructuring the system, I've managed to draw the end-outer curves...
here's a bug I went through in getting it:
^caused by a misordering of the points in the code
^handled the end-outer curves as inner curves by the vert influences
(the line influences don't have any affect unless the line is linked between 2 faces)
I can't render the inner curves...
I'll need my own curve aglorithms to do this
(by then I'll be able to render surfaces)
it would help if I could get feedback info of each vert position in the curve,
then I could render the inner curves and possibly the mesh >_>
in other words...
this is the best I can do with OpenGL's aglorithms. <:/
the only thing that sucks is making the models D:
(verts use lists of 5 slots instead of 3, so I have to model through text)
hopefully I'll be able to build an editor soon >_>
my current model info:
[tt]
points = [
#top square
[-2.0, 2.0, 2.0, 7, 0.5], #top left
[-2.0,-2.0, 2.0, 7, 0.5], #bottom left
[ 2.0,-2.0, 2.0, 7, 0.5], #bottom right
[ 2.0, 2.0, 2.0, 7, 0.5], #top right
#inner square
[-2.0, 2.0, 0.0, 7, 0.5], #top left
[-2.0,-2.0, 0.0, 7, 0.5], #bottom left
[ 2.0,-2.0, 0.0, 7, 0.5], #bottom right
[ 2.0, 2.0, 0.0, 7, 0.5], #top right
#outer square
[-4.0, 4.0, 0.0, 7, 0.6], #top left
[-4.0,-4.0, 0.0, 7, 0.6], #bottom left
[ 4.0,-4.0, 0.0, 7, 0.6], #bottom right
[ 4.0, 4.0, 0.0, 7, 0.6] #top right
]
#[X,Y,Z,R,S]
#R (resolution) is the number of sub-verts in the curve.
#S (strength) is the pull of the inner curve towards the vert
#^
lines = [ #TODO: Normal and UV inputs
#top square
[0,1,0.0], #left
[1,2,0.0], #botom
[2,3,0.0], #right
[3,0,0.0], #top
#TODO: what happens when we reverse p1 and p2?
#top square to inner square
[0,4,0.5], #top left
[1,5,0.5], #bottom left
[2,6,0.5], #bottom right
[3,7,0.5], #top right
#inner square
[4,5,0.8], #left
[5,6,0.8], #botom
[6,7,0.8], #right
[7,4,0.8], #top
#inner square to outer square
[4,8,0.5], #top left
[5,9,0.5], #bottom left
[6,10,0.5],#bottom right
[7,11,0.5],#top right
#outer square
[8,9,0.5], #left
[9,10,0.5],#botom
[10,11,0.5],#right
[11,8,0.5],#top
]
#[p1,p2,S]
#S (strength) is the pull of the outer curve towards the line
#^1.0 yeilds no outer curve while 0.0 yeilds a strait line in the outer curve's position
faces = [ #Quads/Triangles only (for now)
#top square
#[0,1,2,3],
#between top and inner squares
[0,4,8,5], #left
[1,5,9,6], #botom
[2,6,10,7],#right
[3,7,11,4],#top
#between inner and outer squares
[8,12,16,13], #left
[9,13,17,14], #botom
[10,14,18,15],#right
[11,15,19,12] #top
]
#[l1,l2,l3(,l4)]
#index to lines
[/tt]
I've already asked Revel8n and am currently waiting for a responce to see if he'd like to help in developing this... >_>
I'm not sure about asking the blender team, because they don't really like helping anyone with advanced projects...
on the bright side though BJ,
you'll be using one of my implamentations toward my game system, as well as UMC. =D
(my game system is still in the design process... (need a better electronics simulator to manage quaternary))
^my current simulator sucks with control and advanced options >_<
note though that I'm only asking if you'd like to implement it.
the system is nowhere near compleation yet... heh
lol hey BJ... here's an analogy for ya: my gfx system is to normal gfx systems as what Python is to C# :af:
You might be wondering what that has to do with you, but it means I can't actually rotate stuff around.
I can live with that I guess, but I was just wondering if the option to map out different buttons for different commands could ever be a feature at some point?
You might be wondering what that has to do with you, but it means I can't actually rotate stuff around.
I can live with that I guess, but I was just wondering if the option to map out different buttons for different commands could ever be a feature at some point?
While doing some attribute editing, I did run into an odd dilemma. For some reason, whenever I modify the moveset files through BrawlBox, the file size of the PAC file is altered, even though it's fully decompressed. When this dilemma occurs, your game will likely crash if you try to use the file.
Of course, one workaround is to use a hex editor to make the attribute changes. Basically, you first make your changes with BrawlBox, without saving the PAC file, and then export the raw data. You then load the PAC file and the raw data with a hex editor.
In terms of the attribute offsets, when viewing the PAC file, attribute data starts at offset 0x80. The Subspace Emissary attributes start at offset 0x364. Both attribute files have a 0x2e4 file size. Once you've made the changes, save your PAC file, and you'll have the attribute data modified without affecting the PAC file's file size.
Any change to the file will trigger a rebuild. The freeze isn't caused by the filesize, it's caused by character specific nodes and and some articles with either Data3 or certain char specific nodes in them.
Any change to the file will trigger a rebuild. The freeze isn't caused by the filesize, it's caused by character specific nodes and and some articles with either Data3 or certain char specific nodes in them.
Okay. By the way, I figured out what one of the undefined offsets does for the attributes...
0x1F8: This offset apparently affects the default vertical position of the selected character's camera, when the game is paused (NOT in Training).
You notice how Ganondorf's default pause camera position is positioned higher than that of Captain Falcon? That offset is pretty much the reason for this.
Okay. By the way, I figured out what one of the undefined offsets does for the attributes...
0x1F8: This offset apparently affects the default vertical position of the selected character's camera, when the game is paused (NOT in Training).
You notice how Ganondorf's default pause camera position is positioned higher than that of Captain Falcon? That offset is pretty much the reason for this.
Also I'm probably going to fill in most of the floating points for character.txt files. There's loads missing when it comes to known ones, it's slightly annoying when going to edit them. I'll just post them in the thread on kc-mm when I finish.
Okay. By the way, I figured out what one of the undefined offsets does for the attributes...
0x1F8: This offset apparently affects the default vertical position of the selected character's camera, when the game is paused (NOT in Training).
You notice how Ganondorf's default pause camera position is positioned higher than that of Captain Falcon? That offset is pretty much the reason for this.
Also I'm probably going to fill in most of the floating points for character.txt files. There's loads missing when it comes to known ones, it's slightly annoying when going to edit them. I'll just post them in the thread on kc-mm when I finish.
Just a tip, you can rename the values and edit their descriptions right in Brawlbox, and then when you're done you can right click the main moveset node and click "Save Info to Text Files" and that will write them all to the text file for you.
Howdy BJ. Lookin good with Brawlbox lately. =) There is (or I think there's) a problem with BrawlBox0.671. When I went into FitDonkey.pac file, went to the RAW file about his Up-B move, and changed his aerial air resistance to 0.8 and the vertical momentum to 0.72. Ithen went into Brawl, and went into training mode and works fine =). BUT when I go to Vs mode and when it just starts doing the entries of the characters, the game freezes. Do you know what the problem is and how do I fix i?
Howdy BJ. Lookin good with Brawlbox lately. =) There is (or I think there's) a problem with BrawlBox0.671. When I went into FitDonkey.pac file, went to the RAW file about his Up-B move, and changed his aerial air resistance to 0.8 and the vertical momentum to 0.72. Ithen went into Brawl, and went into training mode and works fine =). BUT when I go to Vs mode and when it just starts doing the entries of the characters, the game freezes. Do you know what the problem is and how do I fix i?
Any change to the file will trigger a rebuild. The freeze isn't caused by the filesize, it's caused by character specific nodes and and some articles with either Data3 or certain char specific nodes in them.
btw, I'm working on some file icons for the known (and even unknown) file types.
as well as for invalid data file types such as an mdl0 disguised as a dat
you might like them as they use alot of the images found in win 7's shell icons...
btw, your REFT image resource icon came from Win Media Player 11 :3
(used 'Resource Hacker' to rip the icons and found the image)
Do you think it'll ever be possible to implement the ability to modify vertex points onto BrawlBox? It would seem odd to implement the ability to see vertex points if you can't even edit them.
It's probably a long shot, but I thought it would be something to state.
Would it be possible to have the TopN bone be automatically resized to a FitCharacter's Size attribute? Tested it out with Project M DK. PM DK's Size is 0.915 but the TopN bone in every animation is set to 1, which makes him larger than Project M DK is ingame.
![[TSON]](/data/avatars/l/114/114196.jpg?1360761019){kind=avatar}
