• 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!

Changing Color Effects in Melee!

SNESter

Smash Rookie
Joined
Jan 3, 2015
Messages
9
Location
Texas
NNID
Daysonias
3DS FC
5086-2144-3450
Darkness
(Hyperlinks below are pictures)
Purple and black shine.
Purple and black select hand.
Purple and black re-spawn platform.
Falco's down air now has a darkness effect (also applied it to fox's up-smash).
Clouds turned to purple some mixed with black (if you don't know what I mean refer to Goatlinks posts above).
Purple lasers for both fox and falco.
Illusion/Phantasm changed to purple for fox and falco.
Charging smashes has darkness effect.
Also has the darkness up-b that lance made.
EDIT: Forgot about Marth with Purple and Black Sword Trail (also has darkness effect on down air and neutral B).
-/*\- Download -/*\-

I don't want to double post so...

I was trying to get 990099 into RGBY format, but I must have done something wrong because what I did freezes the game.
Here is the Calculation

990099
R: 153/8 = 19 * 1024 = 19456 >> 4C00

G: 0/8 = 0 * 32 = 0 >> 0

B: 153/8 = 19 >> 13

x4C00 + x0 + x13 = x4C13

x4C13 + x8000 = *[(CC13)]* << I put that in for the lasers and it ended up freezing my game, help would be appreciated.
hey I couldn't find the darkness effect files for falco's and fox's dair/up smash.
 
Last edited:

JoeyS

Smash Rookie
Joined
Jul 23, 2014
Messages
10
I changed Falco's laser color and it worked on Dolphin, but when I play on console it freezes after I try to shoot lasers. I have rainbow shine and it works and I've seen people playing with changed lasers on console, so I don't know what the problem is.

Edit: I changed Pikachu's Thunder and it worked on both, but Falco's still didn't.

Nvm I fixed it.
 
Last edited:

Big D Squirrels

Smash Rookie
Joined
Feb 14, 2015
Messages
10
Location
Santa Cruz, CA
I was requested by @Bennyandthejetz to make a blue and black version of my shine. Here it is. It has everything that was in my original post except with the different shine. https://www.dropbox.com/sh/6ilrkhnnkjr8908/AABElIQ3X5orJzz-xjLKQCDIa?dl=0



I recently made a black and red shine that I'm pretty proud of so I thought I'd stick it up here. I also included that awesome purple firefox/firebird and my purple fox lasers and pale green borderline silver falco lasers. Download link is here: https://www.dropbox.com/sh/ppiv4sak0z9y2tn/AAAIABdPv8fw27tVK3GL9xFoa?dl=0







 

MagicScrumpy

Smash Journeyman
Joined
Feb 25, 2015
Messages
251
Location
Ann Arbor, Michigan
I don't know of anyone who keeps track of what kind of texture is being modified. Are you looking for a specific effect?
I have my player 3 emblems and magnifying glass thing modified to be purple instead of yellow, but the "P3" thing that sometimes appears over the character is still yellow. I don't know where the offset that controls the color of that is. It's not part of the texture.
 

Lanceinthepants

Smash Lord
Joined
Jul 19, 2006
Messages
1,161
Location
Purdue
Scrump, what you are looking for is a value that is calculated and stored in memory (in different places based on certain criteria). It requires ASM coding to change.

Also the value is written again after every stock lost, I can't think of reason the game would do that.
 
Last edited:

AxRandomxMoment

Smash Rookie
Joined
May 15, 2015
Messages
5
Forgive my ignorance but why is that there are several color change effects for Falcons's Falcon Punch, but I haven't seen any for his Falcon Kick?
 

Ray_Rages

Not even mad
Joined
Oct 13, 2014
Messages
17
Location
NEOH

Astaroth

Melee-Fan
Joined
Jan 12, 2014
Messages
42
Location
Bavaria - Germany
Any chance you can edit the blue/black shine file to not effect the fire fox color? I like the blue shine just not the purple fire fox effect.
As far as I know the Shine colors have nothing to do with the Fire Fox colors.
Just look at the FIRST PAGE, Lanceinthepants describes how to change them.
So what you can do is open the file with the edited blue and black shine and open the not edited file next to it and compare them. Then only edit the shine colors in the not edited file. Then you have the normal fire fox and a blue and black shine.
 

Big D Squirrels

Smash Rookie
Joined
Feb 14, 2015
Messages
10
Location
Santa Cruz, CA
Any chance you can edit the blue/black shine file to not effect the fire fox color? I like the blue shine just not the purple fire fox effect.
Unfortunately I'm having compatibility issues on my laptop right now so I don't know if I'll be able to do anything soon. However I will work on it and see if I can get it working again.
 

zankyou

Smash Lord
Joined
Sep 12, 2014
Messages
1,055
Ive been doing some research on making structures in a dat file and Im certain that the 4248 structure is the color node described in the dat documentation. So its a 0x14 length structure were the words are in order of RBBA diffuse, RGBA Ambient, RGBA Specular, unknown (typically 1.0), and unknown (Shininess?).
While I was browsing through the file I found a color node that and 4348 as its 'shininess' value. So this might help you guys in looking for more as they arent necessarily 4248 and its easier to find when you understand a dat file. I didnt look into the other color types. I just came across this one.
 

zankyou

Smash Lord
Joined
Sep 12, 2014
Messages
1,055
Heres something cool.

Now that I understand the dat file looking at masterhand again I can truly appreciate how amazing it is.
This only works for characters but you can change color values here. (didnt test if it was saved but you can at least get the offsets)
The unknowns are the values I described in the previous post.
 

OmnipotentSpoon

Smash Cadet
Joined
Mar 8, 2014
Messages
37
Ive been doing some research on making structures in a dat file and Im certain that the 4248 structure is the color node described in the dat documentation. So its a 0x14 length structure were the words are in order of RBBA diffuse, RGBA Ambient, RGBA Specular, unknown (typically 1.0), and unknown (Shininess?).
While I was browsing through the file I found a color node that and 4348 as its 'shininess' value. So this might help you guys in looking for more as they arent necessarily 4248 and its easier to find when you understand a dat file. I didnt look into the other color types. I just came across this one.
The "unknown (typically 1.0)" is a transparency value. I talked about it here

Okay, so I've been looking into the transparency values that are used for the 42 48 format and think I have it figured out.

So if you don't know what this means, if you have a string like "AA AA AA FF BB BB BB FF CC CC CC FF DD DD 00 00 42 48" the A's are the first color of the format, the B's are the second color, the C's are the third color (which are almost always FF for color effects because they affect the reflected color, it'll be different in some of the character skin files, if there is interest I can write a more complete guide now that I don't have class) and the D's are the transparency value.

What I've learned is pretty simple, the D's are what affect the transparency value, the 0's don't seem to change anything. We knew this already but I haven't seen anything about what ranges were acceptable, probably because in the files they are always either "00 00" or "3F 80". I've found a few things:
  • The D values can be anything without crashing (at least in my testing on mewtwo and mario).
  • If the DD DD value is between 00 00 and 3C 80 there is no change in the effect
  • If the DD DD value is between 3C 80 and 3F 80 the opacity will increase
  • Between 3F 80 and 7F 80 the opacity stays the same
  • Once 7F 81 is reached the transparency reverts to 00 00, and stays that way until FF FF
So basically, if you want to adjust the transparency you have to set it somewhere between 3C 80 and 3F 80. It makes things a lot easier when looking for an effect that has it set to 00 00 initially because when it is set to max opacity the effect usually appears as a bright rectangle/square.

I made an album with results visually in order to show how it works in better detail. The ones that are labeled with only to numbers, "30" or whatever, the second set of D's are both 00, so 30 00 in the previous example.
This is the entire album: http://imgur.com/a/v9u0P#0

A more basic look is given by these two images:

Transparency value of 3F 80

Transparency value of 00 00

Also, does anyone want a guide on how I find color effects? It's kinda just trial and error changing stuff in the .dat files but I've got it down to a science at this point.
Every time that I've looked at changing the 42480000 value I haven't noticed a difference when changing it. I've been planning on checking it since I found a bunch of them with different values in the GrPs.usd file. Most of them are around 42480000 however. An example is at 0x8A60: 191919FF 383E45FF FFFFFFFF 3F800000 4182E8BA.
 
Last edited:

zankyou

Smash Lord
Joined
Sep 12, 2014
Messages
1,055
The "unknown (typically 1.0)" is a transparency value. I talked about it here



Every time that I've looked at changing the 42480000 value I haven't noticed a difference when changing it. I've been planning on checking it since I found a bunch of them with different values in the GrPs.usd file. Most of them are around 42480000 however. An example is at 0x8A60: 191919FF 383E45FF FFFFFFFF 3F800000 4182E8BA.
Dunno know what to tell you. The color node I made worked fine when changing the values. I believe there is a max and min so try not going over 50ish.
 

Gehis514

Smash Apprentice
Joined
Jul 25, 2015
Messages
119
Switch FC
SW 4004 9260 8249
Haven't been able to spend as much time on it recently, but here's the effects I found for Bowser. They're for his neutral, side, and up B.


EfKpData.dat
070707
14D54 - 07 07 07 04 FF FF FF 00 00 7F FF - Side B
Color 1 - Central zigzag slash of side B
Color 2 - Background part of color 1
14E20 - 07 07 07 04 B2 E5 FF 00 00 00 FF - Side B
Color 1 - Central part of aura effect centered on slash of side B
Color 2 - Outer part of the effect
14EE0 - 07 07 07 04 FF FF FF 00 00 7F FF - Side B
Color 1 - Upper zigzag slash of side B
Color 2 - Background part of color 1
14FA0 - 07 07 07 04 FF FF FF 00 00 7F FF - Side B
Color 1 - Lower zigzag slash
Color 2 - Background part of color 1
15E50 - 07 07 07 04 FF FF FF 00 00 4C FF - Up B
Color 1 - Zigzag on up B
Color 2 - Fringe of effect
15F10 - 07 07 07 04 FF FF FF 00 00 7F FF - Up B
Color 1 - Zigzag around bowser during up B
Color 2 - Fringe of effect
15FDC - 07 07 07 04 CC E5 FF 00 00 00 FF - Up B
Color 1 - Effect around bowser during up B
Color 2 - Fringe of effect
1609C - 07 07 07 04 FF FF FF 00 66 B2 FF - Up B
Color 1 - Effect around bowser during up B
Color 2 - Fringe of effect
1615C - 07 07 07 04 FF FF FF 00 00 99 FF - Up B
Color 1 - Effect around bowser during up B
Color 2 - Fringe of effect
1621C - 07 07 07 04 FF FF FF 00 00 7F FF - Up B
Color 1 - Effect around bowser during up B
Color 2 - Fringe of effect
170D4 - 07 07 07 04 FF FF 00 00 FF 4C 00 - Up B
Color 1 - Rays below bowser during ground up B
Color 2 - Part of effect near end of the ray

CF ##
3E0 - CF 00 FF FF FF FF AD DF 00 FF E8 51 - Neutral B
Color 1 - Inner part of fireball near start of B move
Color 2 - Outer part of fireball
3F9 - CF 05 FF FF FF FF AD DF 05 00 00 00 - Neutral B
Color 1 - Outer part of fireballs from neutral B, at the start of it
Color 2 - Inner part of the fireballs
40D - CF 0A FF D9 B8 FF AD DF 0A 00 00 00 - Neutral B
Color 1 - Inner part of fireballs from neutral B, at the end
Color 2 - Outer part of the fireballs
429 - CF 0F 42 42 42 00 AD DF 0F 00 00 00 - Neutral B
Color 1 - Inner part of fireballs at very end of neutral B
Color 2 - Outer part of the fireballs
48E - CF 00 FF FF FF 00 AD DF 00 C0 64 01 - Neutral B
Color 1 - Inner part of faint effect near bowsers mouth during B
Color 2 - Outer part of it
49B - CF 05 FF 80 4A FF AD DF 05 94 25 14 - Neutral B
Color 1 - Inner part of the 48E effect near middle of firebreath
Color 2 - Outer part of the effect
4B9 - CF 03 FF A9 73 FF AD DF 03 4A 27 19 - Neutral B
Color 1 - Inner part of the previous effect at the end of firebreath
Color 2 - Outer part of the effect
4CF - CF 03 FD FD FD FF AD DF 03 00 00 00 - Neutral B
Color 1 - Inner part of another fireball effect
Color 2 - Outer part of the effect - Neutral B
4EC - CF 07 7F 7F 7F 00 AD DF 07 00 00 00 - Neutral B
Color 1 - Inner part of the end of the previous effect
Color 2 - Outer part of the effect
564 - CF 00 FF C6 07 00 AD DF 00 FF 02 06 - Neutral B
Color 1 - Faint circular effect near start of neutral B
Color 2 - Fringes of effect
571 - CF 0C FF EF 40 1E AD DF 0C FF 0A 0E - Neutral B
Color 1 - The central part of the previous effect in the middle of firebreath
Color 2 - The fringes of the effect
585 - CF 0A FF FF FF C8 AD DF 0A 00 00 00 - Neutral B
Color 1 - The central part of the previous effect at the end of firebreath
Color 2 - Fringes of the effect
599 - CF 0A 00 00 00 00 AD DF 0A 00 00 00 - Neutral B
Color 1 - Effect at the very end of firebreath that's only there for like 1 frame
Color 2 - Fringe of the effect
5EC - CF 00 FF FF FF C8 AD DF 00 FF E8 51 - Neutral B
Color 1 - Central part of effect at the very start of firebreath
Color 2 - Fringes of the effect
607 - CF 06 FF FF FF C8 AD DF 06 00 00 00 - Neutral B
Color 1 - More fireballs in the middle of firebreath
Color 2 - Outer parts of the fireballs
626 - CF 0F FF D9 B8 00 AD DF 0F 00 00 00 - Neutral B
Color 1 - Central part of the previous effect at the end of firebreath
Color 2 - Fringes of the effect

4248
14E90 - 80 80 80 FF FF FF FF FF FF FF FF FF 00 00 00 00 42 48 - Side B
Color 1 - Slightly changes color 2
Color 2 - Changes the color of aura effect in side B
As a note, there's a bunch of CF## effects at the start of the EfKpData.dat file that I have no idea what they are.
I've tried to replicate the same flame effects, but there always seems to be a small portion of the original flame around the middle, which looks bad with the blue flame. Any reason why this might happen? Followed the documentation and it's pretty much my first time messing with color effects btw
 

zankyou

Smash Lord
Joined
Sep 12, 2014
Messages
1,055
Last edited:

Tcll

Smash Lord
Joined
Jul 10, 2010
Messages
1,780
Location
The Gates of Darkness
NNID
Tcll5850
I just have to question why it's Y and not A :p
lol

I have a full function for converting this written out in my HAL_Labs DAT script found on that page: ;)
Code:
        def color(fmt,offset=None,alpha=1):
            global DL_size #TODO: remove to increase speed (faster if we don't redefine this function)
            if offset==None: #Direct format
                if fmt==0:DL_size-=2; D=bu16(); return [int(((D>>11)&31)*(255/31)),int(((D>>5)&63)*(255/63)),int((D&31)*(255/31))] #RGB565
                if fmt==1:DL_size-=3; return bu8(['']*3) #RGB8
                if fmt==2:DL_size-=4; return bu8(['']*4)[:3] #RGBX8
                if fmt==3:DL_size-=2; RG,BA=bu8(),bu8(); R,G,B,A=RG>>4,RG&15,BA>>4,BA&15; return [(R*16)+R,(G*16)+G,(B*16)+B,(A*16)+A] #RGBA4
                if fmt==4:DL_size-=3; D=bu24(); return [(D>>18)*(255/63),((D>>12)&63)*(255/63),((D>>6)&63)*(255/63),(D&63)*(255/63)] #RGBA6
                if fmt==5:DL_size-=4; return bu8(['']*4) #RGBA8
            else: #indexed format
                return [255,255,255,255] #yet to be seen (returning white)
usage is simple enough :)

ugeSetColor(None if I=='' else color( attr[3], I, attr[2] ), channel)
ignore the I part :p

vertex color indecies have yet to be seen, but please contact me if you find them :)


EDIT:
btw, the CP attributes (`attr` in my code) describe the format and such.
`I` is `None` if the CP type is GX_Direct (1), otherwise it's the bu8() or bu16() index if not `''` for GX_NULL (0)

normally GX_Direct is used for vertex colors in the DAT format, which can be seen in the images above or in the OP, and is also found in PlCaNr.dat (Captain Falcon)
 
Last edited:

zankyou

Smash Lord
Joined
Sep 12, 2014
Messages
1,055
Just found that the 070707 format is referenced by the unknown structure at the end of the texture structure. Can now probably add cI8 images and color them. Possibly.
I assume that random offsets are just one of the other formats but with changed flags. Now I just have to run across the cf format.
 
Last edited:

Tcll

Smash Lord
Joined
Jul 10, 2010
Messages
1,780
Location
The Gates of Darkness
NNID
Tcll5850
finally got an image to post about the unk1 structure:



Mat Colors
Material
Image Header
Unk 1
Texture
 
Last edited:

DRGN

Technowizard
Moderator
Joined
Aug 20, 2005
Messages
2,179
Location
Sacramento, CA
finally got an image to post about the unk1 structure:



Mat Colors
Material
Image Header
Unk 1
Texture
Actually, I just noticed an error or incomplete detail in this.

Screenshot from 2015-08-20 20-12-42 02.png


The 00055860 in blue above is a part of the texture image data header, the final pointer to the actual image data. Same with the places in red after that.

What do you mean by the 'Texture' areas? What exactly is that data?
 
Last edited:

zankyou

Smash Lord
Joined
Sep 12, 2014
Messages
1,055
Actually, I just noticed an error or incomplete detail in this.
The 00055860 in blue above is a part of the texture image data header, the final pointer to the actual image data. Same with the places in red after that.

What do you mean by the 'Texture' areas? What exactly is that data?
I think he just has it highlighted so that different types of pointers are different colors and the image pointers are just a light color but still highlighted. The texture area is the yellow. Points to image structure and pallets and what I realize a bit ago was the 070707 format. And you can move an image around on the face i think if you play with some of the settings.
 
Last edited:

Tcll

Smash Lord
Joined
Jul 10, 2010
Messages
1,780
Location
The Gates of Darkness
NNID
Tcll5850
the Texture structs contain pointers to the Image Header and optionally Pallet Header structs.
you're right about the pointers you circled (in the Image Headers) pointing to the raw image data.

as for the Texture structs, they're rather a bit complex... refer to my DAT documentation for these:
http://wiki.tockdom.com/wiki/HAL_DAT_(File_Format)#----_Texture_Structures

EDIT:
I think he just has it highlighted so that different types of pointers are different colors and the image pointers are just a light color but still highlighted. The texture area is the yellow. Points to image structure and pallets and what I realize a bit ago was the 070707 format. And you can move an image around on the face i think if you play with some of the settings.
also yes, every valid pointer is highlighted with a light green color, however the overlay blending of that color does not play nicely with HexEdit and that's why they're blue for materials
(the pointers are the first thing highlighted before any structs)

also, @ Z zankyou , the unk1 blocks are not pallet headers, be careful not to confuse others ;)

EDIT2:
also take note the 1 byte highlight at the start of each struct.
that's only meant to show just that, where the previous struct ends and the next struct begins.
(not always accurate as some structs have padding)
 
Last edited:

Lykon

Smash Cadet
Joined
Dec 9, 2012
Messages
66
Location
Monterrey, Mexico

Akli LT

Smash Rookie
Joined
Oct 29, 2015
Messages
13
so. I changed marth colors on his sword swing. but are these changes allowed in tournement play. since u changed the caracter file? in this case "PiMs.dat"
 

RushSSBM

Smash Cadet
Joined
Jan 8, 2014
Messages
46
Location
Germantown, Maryland
Does it though, what TO is not going to allow texture changes?

I'm gonna say yes it's completely fine to use vanilla melee with texture changes in tournament

Maybe not on GIMR's stream but even 20XX is played a lot in tournament, because it's generally fine
 
Last edited:
Top Bottom