xboxscene.org forums

Pages: 1 2 3 [4] 5

Author Topic: Font Exploit Analysis  (Read 486 times)

Wooger

  • Archived User
  • Sr. Member
  • *
  • Posts: 259
Font Exploit Analysis
« Reply #45 on: March 24, 2004, 08:06:00 AM »
QUOTE (ldots @ Mar 24 2004, 06:03 PM)
The current Mech fonts do a lot of patching in memory. What I would hope would be possible was to cut out all this pathing (except the patching of the key), thereby making the exploit more slim in the hope that the exploit would run faster. So this is different from the original Mech fonts in that no in memory patching is needed.

Let me describe again. The original Xbox.xtf and Xbox Book.xtf are renamed to Xbox.bak and Xbox Book.bak. Bert.xtf and Ernie.xtf are present.
xboxdash loads bert and ernie. Patches key and launhes msdash.xbe, hexed to look for the *.bak fonts, and to show Phoenix or whatever in the live tab. PBL could then be renamed to xonlinedash.xbe.

Unfortunately I don't think it will work  sad.gif . I guess it would only work if launching a new app (msdash.xbe) will kill the thread that will eventually check whether the clock is set or not.

Honestly whats the point if the dash in memory is going to look for the original renamed fonts why then would you be concerned if the clock check finishes first.  If it does big whoop cause the dash will then load the original fonts and bring up the clock fix screen...

Which is what the mech fonts do...

Wooger   blink.gif
Logged

ldots

  • Archived User
  • Hero Member
  • *
  • Posts: 822
Font Exploit Analysis
« Reply #46 on: March 24, 2004, 08:12:00 AM »
Wooger you misunderstand me. I would not be doing the xtf -> bak patching in memory in this scenario - that's the original Mech exploit! If the xtf -> bak patching is done succesfully there is no point in loading a hexed msdash right!
Logged

Wooger

  • Archived User
  • Sr. Member
  • *
  • Posts: 259
Font Exploit Analysis
« Reply #47 on: March 24, 2004, 08:14:00 AM »
QUOTE (ldots @ Mar 24 2004, 06:12 PM)
Wooger you misunderstand me. I would not be doing the xtf -> bak patching in memory in this scenario - that's the original Mech exploit! If the xtf -> bak patching is done succesfully there is no point in loading a hexed msdash right!

Whats the point of changing the original dash to look for its fonts with different names then?

Wooger
Logged

Wooger

  • Archived User
  • Sr. Member
  • *
  • Posts: 259
Font Exploit Analysis
« Reply #48 on: March 24, 2004, 08:18:00 AM »
QUOTE (ldots @ Mar 24 2004, 06:12 PM)
Wooger you misunderstand me. I would not be doing the xtf -> bak patching in memory in this scenario - that's the original Mech exploit! If the xtf -> bak patching is done succesfully there is no point in loading a hexed msdash right!

I understand that your saying all you want to do is patch the private key and then from there anything with that key can be run so in the event of a clock loop situation it wouldnt matter the msdash would still function...

Correct?

Wooger
Logged

ldots

  • Archived User
  • Hero Member
  • *
  • Posts: 822
Font Exploit Analysis
« Reply #49 on: March 24, 2004, 08:24:00 AM »
The point would be that if this all worked we would end with public key in memory and a fully booted msdash with a custemized live tab - Just like the original Mech fonts. The only purpose of all this was to cut down on the exploit code in Ernie. But again I dont think this will work.
Would a reloaded msdash end in the same place in memory? If not the "clock thread" wouldn't get to the correct clock setting code of msdash...
Logged

GeToChKn

  • Archived User
  • Full Member
  • *
  • Posts: 245
Font Exploit Analysis
« Reply #50 on: March 24, 2004, 11:25:00 AM »
Well 5101 kernal here and the one i did last night seems to be stuck in a loop that it's not coming out of with or without a network cable plugged into it.
Logged

ldots

  • Archived User
  • Hero Member
  • *
  • Posts: 822
Font Exploit Analysis
« Reply #51 on: April 08, 2004, 03:00:00 AM »
Hi there Grospolina!
Any news on the hybrid fonts? Any discoveries on if and why the clock-setting code can be kernel dependant?
Logged

anu|b|iss

  • Archived User
  • Sr. Member
  • *
  • Posts: 320
Font Exploit Analysis
« Reply #52 on: April 08, 2004, 04:20:00 AM »
Installed the 7/25 fonts on K:5101 D:4920
never had a clock loop once. It comes up with the date/time and freezes after a few seconds, then I reboot and it's set to Jan 1 2001 and I'm good to go.
Logged

Chicken Scratch Boy

  • Archived User
  • Hero Member
  • *
  • Posts: 1054
Font Exploit Analysis
« Reply #53 on: April 08, 2004, 05:25:00 AM »
anu|b|iss:

1. these fonts werent around in 7/25
2. read the thread
3. we dont care
4. your sig is too long
Logged

Neilor

  • Archived User
  • Full Member
  • *
  • Posts: 204
Font Exploit Analysis
« Reply #54 on: April 08, 2004, 06:50:00 AM »
And he's posted the same thing on another thread
Logged

anu|b|iss

  • Archived User
  • Sr. Member
  • *
  • Posts: 320
Font Exploit Analysis
« Reply #55 on: April 08, 2004, 06:56:00 AM »
1. ok, was unaware, always modded with a chip before
2. I had read most of it but skipped to the end to see the current convo (my bad)
3. Nobody said you had to.
4. Fixed that just for you.


5. Neilor, so what?
Logged

Dolfhin

  • Archived User
  • Sr. Member
  • *
  • Posts: 351
Font Exploit Analysis
« Reply #56 on: April 08, 2004, 07:36:00 AM »
Any news yet? I'm looking forward to a new version of your Fonts.
Logged

coltxL2717

  • Archived User
  • Full Member
  • *
  • Posts: 170
Font Exploit Analysis
« Reply #57 on: April 15, 2004, 07:52:00 PM »
Well, I know that this same question has been asked in this same thread, but no one answered.  I am running a 5530 kernel and using the bigfonts25 to exploit my xbox.  In the begining of your thread you said that the Bert Is Cheating On Ernie fonts patched the kernel to allow F: drive support.  Will this also work on my kernal (5530).

A few more questions:

1. How hard is it to splice 2 fonts together?  
2. Why do the Mechfonts refuse to work on Kernel 5530?
Logged

Grospolina

  • Archived User
  • Full Member
  • *
  • Posts: 182
Font Exploit Analysis
« Reply #58 on: April 16, 2004, 05:58:00 AM »
sad.gif

Also, I'm preoccupied with emulators right now. wink.gif
Logged

Zak0

  • Archived User
  • Newbie
  • *
  • Posts: 29
Font Exploit Analysis
« Reply #59 on: August 26, 2003, 09:16:00 AM »
QUOTE
At the beginning of the exception code, it first modifies the instruction before it to cause an infinite loop. That way, if another exception is generated, the Xbox will hang instead of trying to run the exploit again.

After step 6 (above), it checks a flag in the data area. If the flag is set, it hangs the Xbox. If not (default), it proceeds. I'm not sure what use this has.
I'm currently using bigfonts to launch the Phoenix BIOS loader on my box...  About one in three times I boot the box, it locks up before the Phoenix logo appears.  From the work you've done so far, do you think it's possible to modify bigfonts to remove this lock-up-on-purpose behavior?  Or do you think that's even my problem?

All I know is it's pretty annoying to have to reboot the box multiple times...

Thanks!
-Zak
Logged
Pages: 1 2 3 [4] 5
« previous next »