pointer search development

[amoot329]

Err what you really need to do is search for the address of where the spray value is located. Once you have that, you make a dump. Move onto the next level in the game, again locate the address for where the spray value is. Make a 2nd dump.

Once you have both addresses and dumps, you then use the pointer search program to find the pointer.

[Blaster99]

Hi,

there seems to be an error in the pointer searcher tool.

It generates the output
_L 0x00002001 0x00000058
instead of
_L 0x00020001 0x00000058

So it it not possible to do a 16bit or 32bit pointer with the generated output.

Bye
Blaster99

[weltall]

can you give the test case entirely?

[Blaster99]

Hi,

i used cwcheat pointer searcher to search the pointers for naruto ultimate ninja heroes 2.
When i click "make cheat", the generated lines for 16bit and 32bit are not generated correctly.
Compared to the cwcheat wiki -> psp codes , the value for the size of the pointer is not at the right position.

I generated a 32bit pointer:

 
SEARCH 3:
    ...
    ...
RESULT 3:
        Found a pointer at 0x462E90
        It points to 0x11EC640
        Offset from the pointer base is 0x58
    RESULT 4:
        Found a pointer at 0x6115E4
        It points to 0x11EC5E0
        Offset from the pointer base is 0xB8
    RESULT 5:
        Found a pointer at 0xAAB230
        It points to 0x11EC640
        Offset from the pointer base is 0x58
 
calculated cheat:
_L 0x60462E90 0x00003F80
_L 0x00002001 0x00000058

The cweat wiki says:

 
32-bit write 0x6aaaaaaa 0xvvvvvvvv
                 0x0002nnnn 0xiiiiiiii

The pointer searcher has to generate 0x00020001 and not 0x0002001.

Hope you understand what i mean

Bye
Blaster99

[weltall]

ok acknowledged... but why no one noticed it before :S

[Blaster99]

Hi weltall,

i have some requests for your pointer searcher tool. It would be nice if you are adding these functions:

- ability to search in more than 2 dumps at once
- creating more than 1 pointer code at once
- the last used pointer size and value should be saved in the create-cheatcode-form

thx for your time and great software!

Bye
Blaster99

[weltall]

- it increases the complexity of the algorithm and makes the search slower. wouldn't it be better choosing the third dump in one of the slot and do another search?
-why? could you write how it should work? a check list?
- i'm not going to work with windows registry for this so if i'd implement it i'd use a file in the cwd or an hidden file in the home directory.

[Blaster99]

Hi,

oh, did not know that it would increase the complexity.
Maybe a command-line feature is easier to realise ?

Some type of checklist for the creation of the pointer codes would be nice.

Ahh.. i did not mean to save it anywhere.
At the moment, when i create a pointer code, i insert the value and select size of the pointer to create.
When i create the next pointer (maybe because 2 pointer left), i have to insert the value and the pointer size again, although it may be the same size/value.

Bye
Blaster99

[weltall]

Originalmente inviato da Blaster99

Hi,

oh, did not know that it would increase the complexity.
Maybe a command-line feature is easier to realise ?

Some type of checklist for the creation of the pointer codes would be nice.

Ahh.. i did not mean to save it anywhere.
At the moment, when i create a pointer code, i insert the value and select size of the pointer to create.
When i create the next pointer (maybe because 2 pointer left), i have to insert the value and the pointer size again, although it may be the same size/value.

Bye
Blaster99

-sure both in the gfx code and the actually function doing the search.
As for the first the main problem is the windows port: the gtk port for windows doesn't seem to support glade3 (which is based on xml to generate the base window) but only glade2 so it's quite crappy to mantain and i'd like to avoid doing much changes in the gui
still a comand line solution isn't a good idea because there are two ways to implement them in the backend (this affects also the gui case in addition to what i talked about before). The first is doing all possible combinations retaining the current code, this would slow down the process evidently and it's the same you can do now manually by just changing the file, i'd exclude the second way as it's a complete rewrite to check a variable number of files at the same time.

-well i'll see.

-ah you meant keeping the selected code type in the generation window. this would require a global variable or a variable being sent and received from the function calling it as that window is destroyed when going out from it or it can be fixed with the check box too.

[Blaster99]

Originalmente inviato da weltall

... The first is doing all possible combinations retaining the current code, this would slow down the process evidently and it's the same you can do now manually by just changing the file, i'd exclude the second way as it's a complete rewrite to check a variable number of files at the same time.

-ah you meant keeping the selected code type in the generation window. this would require a global variable or a variable being sent and received from the function calling it as that window is destroyed when going out from it or it can be fixed with the check box too.

- Hmm.. why is it slower ? If i select the maybe 4 dumps in the pointer-searcher. Then it compares dump1-dump2 keeping same pointers (same as now) and then automatically compares dump3-dump4 with the found pointers(same as now).
Would it not be theoretically the same but without selecting the new dumps after each search ?

- Yes, a global variable would be a solution.

- I tried to make a perl script for searching the pointers for using it in a batch file, but i was not able to find out for what your pointer searcher searches for to find the pointers *g* Maybe you can give me a hint.

thx for taking your time for my feature requests

bye
Blaster99