Important Melee, Hacks, and You -- New Hackers Start Here, in the OP!

[Tcll]

@ Z zankyou : may I ask why you post attachments
there's a thing called [ img ] tags... lol
(I try to avoid attachments because they get uploaded here)

epic job though
although I'm sure that little Z-order issue around the edges of the leaves can be fixed.

 

[Itaru]

oh you're using dev4.5 (the sucky-cruddy version... lol)
Ctrl+I (as stated in the README)

I've stopped distributing that version though since it doesn't do anything properly... lol
(I do mean that, after designing 3.0 and rebuilding 3.0a, there's quite alot I was doing wrong in dev4.5)
^ dev5 isn't all that glorious either, it still can't quite do animation just yet, and the format is still a mess.

I don't quite have a release for dev5 yet, there's still quite alot of work that needs to be done on it...
however, you can download my actively updated repo here:
https://copy.com/DB24NZi4ejlxFo4I
(on windows: run the bat unless you actually see the exe)

just stay away from SIDE for now... lol
I'm trying to work on some intellisense/code-insight for the code editor...
I still have to figure out how to get the FS to cooperate through the fake 3.0 backend


as of this post though, don't download just yet, I'm reworking the transformer to get rid of the common Singular Matrix error I keep getting.
I'll edit this post when I finish to let you know it's safe to download.

EDIT:
well, everything's working properly now, and I've fixed half of the singular matrix issues... but the other half is in the verification...
gonna do a few edits to the GUI real quick...
- A-Z directory sorting for the file browser
- removing the model-export path's '...' from the top-left corner of the screen

Thanks!!!!

 

[zankyou]

@ Z zankyou : may I ask why you post attachments
there's a thing called [ img ] tags... lol
(I try to avoid attachments because they get uploaded here)

epic job though
although I'm sure that little Z-order issue around the edges of the leaves can be fixed.

I actually think its because I made all of the color nodes white. But if its not then its a small detail that I probably cant fix. Right now I gotta figure out why images for the next part of the mesh Im trying to import are on the wrong face.

 

[RegalSwan]

How would I go about porting SD Remix' characters into 20XX? It's probably a question with an obvious answer, I really am sorry, but I just want to make a ROM with all of SDR's character in 20XX so I can use their features in the Smash club I'm starting locally.

 

[zankyou]

Made some progress. Found the flags responsible for normals. Now the tree and grass stuff render properly.

Guys if you want to help with figuring out how the dat file works look in http://smashboards.com/threads/dat-file-research.417755/. Its really as easy as copying over other unk flags from other similar structs and noting the difference. Not hard but takes forever when its just one person doing.

 

[MagicScrumpy]

How would I go about porting SD Remix' characters into 20XX? It's probably a question with an obvious answer, I really am sorry, but I just want to make a ROM with all of SDR's character in 20XX so I can use their features in the Smash club I'm starting locally.

Just put the files for both 20XX and SDR in the same root

 

[zankyou]

Just put the files for both 20XX and SDR in the same root

Its not that easy. SDR uses the fsm engine which has been ported over to 1.02 by achilles but its probably buried in post. Not sure if Achilles made that version compatible with SDR but otherwise it cant be done without knowing wheres a safe spot to inject the fsm engine.

 

[MagicScrumpy]

I forgot that SDR did that lol. My bad. I'll look for Achilles's FSM post real quick.

 

[Doq]

Its not that easy. SDR uses the fsm engine which has been ported over to 1.02 by achilles but its probably buried in post. Not sure if Achilles made that version compatible with SDR but otherwise it cant be done without knowing wheres a safe spot to inject the fsm engine.

Crazy Hand can do FSM on a 20XX image.
It's just a matter of finding out every single FSM entry in SDR

 

[Tcll]

OMG what have they done to SmashBoards, it looks disgusting now D:

sorry for OT...
any progression with materials??

EDIT:
btw fixed up the material structure a tad (on my format page)
those 4 bytes in the structure at 0x40 are really 2 bu16 values

unk_flags = bu16() # 0x002X enables alpha, but disables it when 0x006X
unk_value = bu16() # seems to play a part in modifying the specular dispersion, like shininess

^ btw, if you can understand that, you can write a UMC-script
just thought I'd make a point... lol

 

[Air Bair]

Does anybody know what the offset is for [31 BGBu -> txt_0088_14.tga] in Steelia's MnSlMap.usd texture pack? The placements file only includes 1-29, so i'm not sure how to find the other ones =\

 

[DRGN]

lol I've seen both terms, I picked up Pallet first before finding out years later it was really Palette
so I have implementations for both.
hey Kaz yelled at me for this as well... haha

and for reading (not yet writing) GX texture data, I've written my own lib "RVL_IMG" for handling that stuff.
however, for handling PNG (among nearly 50 other formats), I'm using Pillow, although palette support is borked because pillow devs don't understand how to document the supported formatting... heh

and yea, data length is determined by (width*height)*data_size
unless it's RGBA8 which is 32bytes RG and 32bytes BA tor each texel
and also unless it's a special format like CMPR which is 8 bytes for a 4x4 texel

As I mentioned before, that method for calculating data length doesn't always work. At least, not if you're using the actual display dimensions of the texture. I'm pretty sure it's because there are some pixels that fall outside of the visible range. Thus they add to data size, but aren't accounted for in any recorded dimensions such as in dat file data headers or tpl files. This is just theory atm, based on the Mario Kart wiki page on image formats. But if you want, I can post again when I get around to writing the correct calculation.

Does anybody know what the offset is for [31 BGBu -> txt_0088_14.tga] in Steelia's MnSlMap.usd texture pack? The placements file only includes 1-29, so i'm not sure how to find the other ones =\

Check my thread for methods on finding textures. Here's some that I've found:

30 - 0x2cd40
31 - 0x880
32 - 0x800
33 - 0x2880
34 - 0x38840
35 - 0x6a740
36 - 0x36340
37 - 0x2dd40
38 - 0x2d540
39 - 0x2d940
40 - 0xdcc0

 

[zankyou]

As I mentioned before, that method for calculating data length doesn't always work. At least, not if you're using the actual display dimensions of the texture. I'm pretty sure it's because there are some pixels that fall outside of the visible range. Thus they add to data size, but aren't accounted for in any recorded dimensions such as in dat file data headers or tpl files. This is just theory atm, based on the Mario Kart wiki page on image formats. But if you want, I can post again when I get around to writing the correct calculation.



Check my thread for methods on finding textures. Here's some that I've found:

30 - 0x2cd40
31 - 0x880
32 - 0x800
33 - 0x2880
34 - 0x38840
35 - 0x6a740
36 - 0x36340
37 - 0x2dd40
38 - 0x2d540
39 - 0x2d940
40 - 0xdcc0

Hmm, I thought it was based on block size more so than pixel size. Like you would round the block up and use that as your size.

 

[Tcll]

As I mentioned before, that method for calculating data length doesn't always work. At least, not if you're using the actual display dimensions of the texture.

if the format is CMPR:
data_size = ( (pixel_width/4) * (pixel_height/4) )*8

note: bytes of a 16 pixel (4x4) texel:
0x0000 0000 00000000

RGB565 color1
RGB565 color2
pixel-map

if the format is I4 or CI4:
data_size = (pixel_width/2) * (pixel_height/2)

if the format is RGB565, RGB5A3 or IA8:
data_size = ( pixel_width * pixel_height )*2

the rest follows a similar pattern to the last (just change the 2 to 4)
I havn't gotten into the CI14x2 format just yet, so I don't know much about it...

also, texel (block) size is just a division of the W or H by 4 or 8 depending on the format.


btw, that's all off the top of my head and untested, so let me know if anything's wrong


EDIT:
corrected CMPR info.

EDIT2:
just realized "texel" was the wrong term as that applies to the pixel of the "tile"
note: the tile is commonly termed "tlut" in code

 

[SinsOfApathy]

also, texel (block) size is just a division of the W or H by 4 or 8 depending on the format.

Really sick of these new reply layouts on mobile. Literally unusable and can't copy-paste.

http://hitmen.c02.at/files/yagcd/yagcd/chap15.html#sec15.35

 

[Tcll]

may as well post this here too... heh
note: this is for programmers, not noobs.
(although don't let that stop you noobs from attempting to figure it out)
^ it's really the same thing as writing a UMC script

wrote something that might help you guys with hacking
https://copy.com/670IF7htQIKplXTJ
^rw.py read/write binary files using a UMC 3.0a dev5-ish data system

when importing:
from rw import *

see the bottom of the script for example usage.

if you have any traceback issues, let me know

EDIT:
later on I'll add some extended support for array()() and struct()()
and even further down the road, possibly pointer support, like UMC 3.0

EDIT2:
btw

Really sick of these new reply layouts on mobile. Literally unusable and can't copy-paste.

http://hitmen.c02.at/files/yagcd/yagcd/chap15.html#sec15.35

if that's in response to me, it's not very helpful
although it does cover formats I may have missed, so thanks

but it doesn't state that RGB565 is 16bit format following: R 5bits, G 6bits, B 5bits, with any technical info on managing those ranges.
same for RGB5A3 and such which is really 2 formats: RGB555 or RGB4A3 depending on the top bit.

it also doesn't state (for noobs) the texels needing to be built before arranged to make the image.
(I was confused on this when I started learning this for my REFT viewer)

quite a lack of needed information there.

again, this is all off the top of my head... lol

EDIT3: (invalidated)
un-recalled memory refresher SinsOfApathy , thanks

CMP(R) format texels being arranged like so:

01 02 |05 06
03 04 |07 08
09 10 |13 14
11 12 |15 16

EDIT4:
nvm, my (RVL_IMG) code invalidates that layout:

    return [p[(di>>30)&3],p[(di>>28)&3],p[(di>>26)&3],p[(di>>24)&3],
            p[(di>>22)&3],p[(di>>20)&3],p[(di>>18)&3],p[(di>>16)&3],
            p[(di>>14)&3],p[(di>>12)&3],p[(di>>10)&3],p[(di>>8)&3],
            p[(di>>6)&3] ,p[(di>>4)&3] ,p[(di>>2)&3] ,p[(di>>0)&3]]

arrangement being:
01 02 03 04
05 06 07 08
09 10 11 12
13 14 15 16

as:
t44(1) t44(2)
t44(3) t44(4)

ultimately a t88 (tile 8x8)

 

[WinterWonter]

What is a character's get up forward/back roll called in CrazyHand? Is there a way to find that?

 

[Achilles1515]

What is a character's get up forward/back roll called in CrazyHand? Is there a way to find that?

Debug menu -> Turn DBLevel to DEVELOP -> Go in a match and Y+D-pad Down to show Action State Names

 

[WinterWonter]

Debug menu -> Turn DBLevel to DEVELOP -> Go in a match and Y+D-pad Down to show Action State Names

Oh duh. I've done this a billion times before with aerials. I just thought this wouldn't apply to every animation. Good to know, thanks!

 

[SinsOfApathy]

if that's in response to me, it's not very helpful
although it does cover formats I may have missed, so thanks

but it doesn't state that RGB565 is 16bit format following: R 5bits, G 6bits, B 5bits, with any technical info on managing those ranges.
same for RGB5A3 and such which is really 2 formats: RGB555 or RGB4A3 depending on the top bit.

http://hitmen.c02.at/files/yagcd/yagcd/chap17.html#sec17.12.1

It's in the documentation.

 

[Tcll]

http://hitmen.c02.at/files/yagcd/yagcd/chap17.html#sec17.12.1

It's in the documentation.

huh... don't remember seeing that section before...
glad they finally updated some time ago (after I gave up). /P'd
I got my info from quite a few other sources.

thanks for the update

 

[DRGN]

Tcll
So I finally sat down to address the image data size bug that's been sitting here forever, and was pleased at how easy/fast it actually was to fix. (The problem with your last suggestion btw was that it assumes the texture's visible dimensions would be even multiples of the block width/height.) I knew some rounding would need to be involved (assuming the problem was actually related to blocks & non-visible pixels, which I was like 98% sure of), so I threw that in there, and here's the finished function:

[import math]
def getImageDataLength( width, height, imageType ): # Arguments should each be ints.
    byteMultiplyer = { # Defines the bytes required per pixel for each image type.
        0: .5, 1: 1, 2: 1, 3: 2, 4: 2, 5: 2, 6: 4, 8: .5, 9: 1, 10: 2, 14: .5 }
    blockDimensions = { # Defines the block width and height for each image type.
        0: (8,8), 1: (8,4), 2: (8,4), 3: (4,4), 4: (4,4), 5: (4,4), 6: (4,4), 8: (8,8), 9: (8,4), 10: (4,4), 14: (8,8) }

    blockWidth = blockDimensions[imageType][0]
    trueWidth = math.ceil( float(width) / blockWidth ) * blockWidth

    blockHeight = blockDimensions[imageType][1]
    trueHeight = math.ceil( float(height) / blockHeight ) * blockHeight

    return int( trueWidth * trueHeight * byteMultiplyer[imageType] )

I've replaced all instances of calculating the image data length with this, and it seems to be working flawlessly for all textures. ^_^

wrote something that might help you guys with hacking
https://copy.com/670IF7htQIKplXTJ
^rw.py read/write binary files using a UMC 3.0a dev5-ish data system

I'm not really sure what this is for. Read/write what kind of binary files?


SinsOfApathy , yeah, that's a good resource, but for textures, these on the Custom Mario Kart Wiki are better:
On the TPL format
On the data formatting

I do notice YAGCD has extra notes on the on the XFB and YCbYCr format though.

Edit:

Hmm, I thought it was based on block size more so than pixel size. Like you would round the block up and use that as your size.

lol yep.

 

[Tcll]

I'm not really sure what this is for. Read/write what kind of binary files?

any kind, it just reads/writes binary data through a provided UMC-ish data system.

it's pretty much a replacement for import struct.
I've always hated the struct module due to it's lack of support and extensive code.

again example usage is at the bottom of the code.

 

[Achilles1515]

Z zankyou , do you ever use RAM “file patching” to do your testing on stage files? This is certainly the most efficient way to test small changes to structures or unknown values or whatever. It involves changing the loaded .dat file contents in the RAM after is has been loaded, but before it has been read and modified.

Set a breakpoint at 80018130 and load a stage. Game will pause at breakpoint. (You’ll most likely want to do a Dolphin save-state right here).

Take pointer in r30 and add 0x10 to it. Go to that address in the RAM and take the pointer at that address and add 0x4. Go to that address in the RAM and it holds the pointer to the start of the stage .dat file.

So if you want to change a value at 0x10000 in the .dat, just add 0x10000 to the .dat start pointer, change the value, then resume game.

 

[Tcll]

I'm still waiting for an FPC...
I can't run Dolphin (can't configure my controllers) and my wii can't read disks, so I'm stuck with CFG which can load GCT files.
I'm not sure how this works with DM, but it works

 

[Achilles1515]

I'm still waiting for an FPC...
I can't run Dolphin (can't configure my controllers) and my wii can't read disks, so I'm stuck with CFG which can load GCT files.
I'm not sure how this works with DM, but it works

Can't you just manually edit the .ini file and set it up to your keyboard?

Dolphin Emulator -> Config -> GCPadNew.ini

Spoiler: My GCPadNew.ini

[GCPad1]
Device = DInput/0/Keyboard Mouse
Buttons/A = A
Buttons/B = S
Buttons/X = Q
Buttons/Y = W
Buttons/Z = I
Buttons/Start = E
Main Stick/Up = UP
Main Stick/Down = DOWN
Main Stick/Left = LEFT
Main Stick/Right = RIGHT
C-Stick/Up = V
C-Stick/Down = M
C-Stick/Left = B
C-Stick/Right = N
Triggers/L = O
Triggers/R = P
D-Pad/Up = `1`
D-Pad/Down = `4`
D-Pad/Left = `2`
D-Pad/Right = `3`
[GCPad2]
Device = DInput/0/Keyboard Mouse
Buttons/A = `5`
Buttons/B = `6`
Buttons/X = Q
Buttons/Y = W
Buttons/Z = I
Buttons/Start = L
Main Stick/Up = `7`
Main Stick/Down = `0`
Main Stick/Left = `8`
Main Stick/Right = `9`
C-Stick/Up = V
C-Stick/Down = M
C-Stick/Left = B
C-Stick/Right = N
Triggers/L = O
Triggers/R = P
D-Pad/Up = `1`
D-Pad/Down = `4`
D-Pad/Left = `2`
D-Pad/Right = `3`
[GCPad3]
Device = DInput/0/Keyboard Mouse
Main Stick/Up = T
Main Stick/Down = G
Main Stick/Left = F
Main Stick/Right = H
Buttons/A = U
Buttons/B = Y
Buttons/X = D
Buttons/Y = R
Buttons/Z = V
Buttons/Start = PERIOD
Main Stick/Radius = 100.000000
Main Stick/Modifier = LSHIFT
Main Stick/Modifier/Range = 50.000000
C-Stick/Radius = 100.000000
C-Stick/Up = K
C-Stick/Down = R
C-Stick/Left = `Axis Z-`
C-Stick/Right = `Axis Z+`
C-Stick/Modifier = LCONTROL
C-Stick/Modifier/Range = 50.000000
Triggers/L = L
Triggers/R = `Button 5`
Triggers/L-Analog = `Slider 0+`
Triggers/R-Analog = `Slider 1+`
D-Pad/Up = `0`
D-Pad/Down = BACK
D-Pad/Left = MINUS
D-Pad/Right = EQUALS
[GCPad4]
Device = DInput/0/Keyboard Mouse
Main Stick/Up = T
Main Stick/Down = G
Main Stick/Left = F
Main Stick/Right = H
C-Stick/Up = Y
D-Pad/Down = B
Buttons/Z = SLASH
Buttons/A = APOSTROPHE
Buttons/B = RBRACKET
Buttons/Start = RBRACKET

 

[Tcll]

why would I wanna do that when I have a GCN controller...
and I'm not smart enough and am too lazy to figure out the buttons... lol

and joytokey doesn't work very well with the joystick range

EDIT:
but even still, I don't trust Dolphin to be stable enough on Xubuntu 14.04

EDIT2:
updated rw.py a tad btw, the example is a little easier to understand
https://copy.com/670IF7htQIKplXTJ
not only that, but I removed the functions and made them global objects so it works more like UMC.

buuut this still doesn't work properly:
>>> print bu32 == bu(4)
False

 

[Achilles1515]

why would I wanna do that when I have a GCN controller...
and I'm not smart enough and am too lazy to figure out the buttons... lol

and joytokey doesn't work very well with the joystick range

EDIT:
but even still, I don't trust Dolphin to be stable enough on Xubuntu 14.04

EDIT2:
updated rw.py a tad btw, the example is a little easier to understand
https://copy.com/670IF7htQIKplXTJ
not only that, but I removed the functions and made them global objects so it works more like UMC.

buuut this still doesn't work properly:
>>> print bu32 == bu(4)
False

Well if you're just being lazy then don't say you can't run Dolphin on your computer.

 

[Tcll]

I'm saying I can't run dolphin because it freezes when I try to configure it.

why should I have to go through extra work just to configure dolphin.
I don't know anything about the ini files and don't really have the time to learn...
why can't the UIs just work like they're supposed to.

EDIT:
I do plan on going to the devs in time and helping them figure out the issue so it doesn't happen to anyone else.
but that's when I get some time away from programming.

I only say I'm lazy because I'm overworked (and it's my fault for being in that position)

EDIT2:
btw, fixed the issue
live test in IDLE shell:

>>>
string(4): magi c
string(u8): 0x07 string1
u32: 16384
s16: -255
f32: 1.0
h16: 0x00 F080
string(): string2 \x00
Press Enter to exit...
>>> u32 == u(4)
True
>>> u32 == bu(4)
False
>>>

^something I need to fix in UMC 3.0a as well... lol

 

[zankyou]

Z zankyou , do you ever use RAM “file patching” to do your testing on stage files? This is certainly the most efficient way to test small changes to structures or unknown values or whatever. It involves changing the loaded .dat file contents in the RAM after is has been loaded, but before it has been read and modified.

Set a breakpoint at 80018130 and load a stage. Game will pause at breakpoint. (You’ll most likely want to do a Dolphin save-state right here).

Take pointer in r30 and add 0x10 to it. Go to that address in the RAM and take the pointer at that address and add 0x4. Go to that address in the RAM and it holds the pointer to the start of the stage .dat file.

So if you want to change a value at 0x10000 in the .dat, just add 0x10000 to the .dat start pointer, change the value, then resume game.

TIME TO TEST SOME FLAGS! (after class)

 

[SinsOfApathy]

I'm saying I can't run dolphin because it freezes when I try to configure it.

why should I have to go through extra work just to configure dolphin.
I don't know anything about the ini files and don't really have the time to learn...
why can't the UIs just work like they're supposed to.

EDIT:
I do plan on going to the devs in time and helping them figure out the issue so it doesn't happen to anyone else.
but that's when I get some time away from programming.

I only say I'm lazy because I'm overworked (and it's my fault for being in that position)

I saw that your issue is reported as #8462.

And I assume you're doing the PPA method and installed the dependencies as seen here now? https://wiki.dolphin-emu.org/index.php?title=Installing_Dolphin#Ubuntu

Your issue shows a different PPA that was outdated shortly after you reported the issue, so wasn't sure.

 

[Tcll]

SinsOfApathy : I'd actually forgot I'd reported it... haha
hadn't heard anything back, so I didn't know if there was an update or if it was fixed or not...
(been too busy to look)
so I'll try it out here soon and see if anything's fixed
thanks.

EDIT:
welp... installed all dependencies and installed the dev-version and it doesn't even load w/o crashing...
will the stable version work just as well for scanning data and such??

EDIT2:
welp... the stable version coughs too:

tcll@tcll-AY589AAR-ABA-a4317c:~$ dolphin-emu
Aborted (core dumped)
tcll@tcll-AY589AAR-ABA-a4317c:~$


EDIT3:
AH!
gcc-4.9-base was not installed... synaptic shows nothing...
well then >_>

EDIT4:
you've got to be kidding me:

tcll@tcll-AY589AAR-ABA-a4317c:~$ sudo apt-get install gcc-4.9-base
Reading package lists... Done
Building dependency tree
Reading state information... Done
gcc-4.9-base is already the newest version.

 

[Tcll]

Achilles1515 is it possible to load Melee 20XX through CFG USB Loader??

I can't get dolphin to work for crap... think I might have to format, reinstall, and upgrade linux again.
this goes to anyone:
start a private convo with me about which linux you think I should try (I'm considering Arch or Mint)
(I will NOT do anything KDE, it's the linux version of Windows 8 (in terms of RAM/CPU))

 

[pmilkosky]

Hi, I'm pretty new to all of this, I started out with crazy hand and have no idea how to use hex edit. If anyone can show me a step by step tutorial on how to edit any projectiles, for example, make mario's fire a stream of fireballs or slow fire a massive fireball, a bunch of newbies and I would be really glad.

 

[Achilles1515]

Achilles1515 is it possible to load Melee 20XX through CFG USB Loader??

I can't get dolphin to work for crap... think I might have to format, reinstall, and upgrade linux again.
this goes to anyone:
start a private convo with me about which linux you think I should try (I'm considering Arch or Mint)
(I will NOT do anything KDE, it's the linux version of Windows 8 (in terms of RAM/CPU))

Yes, just use a 20XX SSBM ISO. I use CFG Loader for when I want to test new Gecko codes on console.

 

[SinsOfApathy]

Tcll

Trying running it through GDB and see if you can get a stacktrace. Bit of an issue for me to install flavor over normal Ubuntu that I run through VirtualBox. I know XUbuntu has a lower footprint, but now that vanilla is back to distributing FFMPEG it's non-issue to me.

 

[Tcll]

Yes, just use a 20XX SSBM ISO. I use CFG Loader for when I want to test new Gecko codes on console.

erm... guess I should've stated myself a bit better... heh
I wanted to know if it would still load of the SD, or if I had to continually patch the ISO
(FAT32 is not very good on linux and usually corrupts almost every time I write to it)

speaking of, know how to format a FAT32 drive to a 32 or 64 byte cluster size on linux??
I'm sick of DM taking forever to load... lol

SinsOfApathy :
alright, I've never used GDB though
what do I have to do??

 

[SinsOfApathy]

SinsOfApathy :
alright, I've never used GDB though
what do I have to do??

$ gdb
(gdb) run ./path/to/dolphin
----Open up Dolphin for the crash
(gdb) bt

*Note: Dolphin is multi-threaded so instead of bt, you may need to run "thread apply all backtrace"

I generally do it with stuff that I'm compiling with gcc -g0 for no optimization, so I don't know what that may or may not mean for Dolphin. We'll have to see.

 

[Achilles1515]

erm... guess I should've stated myself a bit better... heh
I wanted to know if it would still load of the SD, or if I had to continually patch the ISO
(FAT32 is not very good on linux and usually corrupts almost every time I write to it)

speaking of, know how to format a FAT32 drive to a 32 or 64 byte cluster size on linux??
I'm sick of DM taking forever to load... lol

SinsOfApathy :
alright, I've never used GDB though
what do I have to do??

If you're talking about having the Root folder on your SD card, then idk. I just use ISOs.

 

[Tcll]

SinsOfApathy : slightly modified for readability:

tcll@tcll-AY589AAR-ABA-a4317c:~$ gdb
GNU gdb (Ubuntu 7.7.1-0ubuntu5~14.04.2) 7.7.1
Copyright (C) 2014 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word".

(gdb) exec-file /usr/games/dolphin-emu 
(gdb) run /usr/games/dolphin-emu 
Starting program: /usr/games/dolphin-emu /usr/games/dolphin-emu
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffde305700 (LWP 32039)]
[New Thread 0x7fffddb04700 (LWP 32040)]

Program received signal SIGILL, Illegal instruction.
0x00000000009d7188 in ?? ()
(gdb) thread apply all backtrace

Thread 3 (Thread 0x7fffddb04700 (LWP 32040)):
#0  0x00007ffff4f2b12d in poll () at ../sysdeps/unix/syscall-template.S:81
#1  0x00007ffff25a8db4 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#2  0x00007ffff25a90da in g_main_loop_run () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#3  0x00007ffff0ae4e96 in ?? () from /usr/lib/x86_64-linux-gnu/libgio-2.0.so.0
#4  0x00007ffff25cdd55 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#5  0x00007ffff2ca2182 in start_thread (arg=0x7fffddb04700) at pthread_create.c:312
#6  0x00007ffff4f3847d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

Thread 2 (Thread 0x7fffde305700 (LWP 32039)):
#0  0x00007ffff4f2b12d in poll () at ../sysdeps/unix/syscall-template.S:81
#1  0x00007ffff25a8db4 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#2  0x00007ffff25a8ebc in g_main_context_iteration () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#3  0x00007fffde30d1ad in ?? () from /usr/lib/x86_64-linux-gnu/gio/modules/libdconfsettings.so
#4  0x00007ffff25cdd55 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#5  0x00007ffff2ca2182 in start_thread (arg=0x7fffde305700) at pthread_create.c:312
#6  0x00007ffff4f3847d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

Thread 1 (Thread 0x7ffff7f96b00 (LWP 32035)):
#0  0x00000000009d7188 in ?? ()
#1  0x0000000001973fa0 in ?? ()
#2  0x00007fffffffcb30 in ?? ()
#3  0x000000000208caf0 in ?? ()
#4  0x0000000000000020 in ?? ()
#5  0x000000000208cb80 in ?? ()
#6  0x00000000009d7ab0 in ?? ()
#7  0x00007fffffffcb30 in ?? ()
#8  0x000000000208caf0 in ?? ()
#9  0x00007fffffffcb30 in ?? ()
#10 0x00007fffffffcac0 in ?? ()
#11 0xffffffffffffffff in ?? ()
#12 0x00007fffffffcb30 in ?? ()
#13 0x0000000001973fa0 in ?? ()
#14 0xffffffffffffffff in ?? ()
#15 0x0000000001973fa0 in ?? ()
#16 0x00000000009f04c1 in ?? ()
#17 0x00007fffffffcb30 in ?? ()
#18 0x0000000000000000 in ?? ()
(gdb)

Program received signal SIGILL, Illegal instruction.

haha, typo mesees
SIGKILL

Achilles1515 :
yea, I hate operating on my USB (cause I'm scared of FAT32 corruption), so I'd prefer to put the ISO on my USB while managing the hacked files on my SD (like Brawl).
at least if my SD corrupts, I back it up normally after every edit, so there's no loss...

but that's 2GB (SD) vs 32GB (USB)