MAXConvert Beta v0.5

[Vampmaster]

If I've understood you right, shouldn't this just equate to the max command in my last post?:
4aaaaaaa xxxxyyyy
dddddddd 00000000

I get the bit about the 8 and 16 bit versions of the command not working, though. When I was working on the TRANScrypt, SilverSephiropth was talking to me about the maxcrypt taking descriptions and I figured that there's always the chance of someone giving their code a name that looks like a code. I also figured people might want to comment out codes sometimes. It might be usefull to add someting like that in the MAXConvert. Another reason is (I don't know how you've handled this yet->), but if someone puts for example some AR2 codes in the middle of some AR MAX codes, would it decrypt the MAX and then go back and convert the AR2 code to whatever the MAX was being converted to.

[Pyriel]

It does, but if a value increment is specified, some devices don't handle that. CBv7 handles it as:

4aaaaaaa xxxxyyyy
dddddddd iiiiiiii

where "i" is the value increment, but v1-6 don't handle that extra feature. I've been told GSv3+ can handle it and I know AR2/GS2 v1 and v2 cannot. Simulating the effect of the code will require checking the input and possibly exploding the code into a series of single-address writes.

Example:

MAX:
00000000 8596E78C
00001000 01030002

Going to a GSv3, it could be done as:
4196E78C 00030002
00001000 00000001

On a CBv1, it would have to be:
2196E78C 00000001
2196E794 00000002
2196E79C 00000003

When the value increment is absent, yeah, the MAX 32-bit code is equivalent to the code used by all other devices.

No, it assumes that whatever type of input is specified is the only type of code entered. Obviously it's easy to distinguish between encrypted MAX codes and other code types. I'm of the opinion that the other encryptions produce results that are too similar to be accurately separated from one another by a program. Although I haven't really looked into it. CBv7+ encryption produces results that look a lot like ARv1 codes at first glance (to me anyway), for example.

At this point, code recognition would just introduce an extra point of failure that I don't need. I may be tempted to wedge it in later. I think like Parasyte in some ways, though. If you're going to be messing with codes, you'd better be willing to put some thought and effort into it other than copying and pasting and clicking buttons. It's really not too much to ask that you figure out what type of code you're entering.

[Vampmaster]

Well, make sure you check that thing where raw max to raw makes E commands into Fs. It might need to be this in the 32 bit multi skip: outCode[oidx] |= (0x0FFF0000 & (c / 2 - (iidx + 1)) << 16); or have it find the number of codes in the input and break the loop after that.

EDIT: BTW, in my last post I wasn't talking about adding something like Dark Killers code look up program, just the ability to comment out codes.

EDIT 2: The following code didn't encrypt correctly when I tried converting it to AR2v1:

200fff90 10000013
200fff98 34020400
200fff9c 24010024
200fffa0 00010c00
200fffa4 ac221c68
200fffc8 0804bb42
2014073c 0c03ffe5

I think it was just the addresses that were wrong.

[Pyriel]

Yeah, the wrong seeds were hard-coded in. I noticed it a while ago, but that device is so outdated it hardly seemed worth making a new version just for that.

The new version can be downloaded here:
MAXConvert V0.6

It contains numerous additions and bug-fixes. Check the readme for more details (yeah, the seeds for the AR1 encryption are fixed). Some time soon the readme will become an html file with more in depth examples and instructions.

Oh, one thing I forgot to include in the readme. Where it talks about fills that exceed allowable limits, check the All Sphere Break Coins code for FFX-2 on the CJ's site for an example of that. They used the wrong command (16-bit instead of 32) so some of the fills appear to be thousands of bytes wide. I can't imagine a situation where you would actually need to fill more than 1,000 bytes with a particular value, hence the limit. At minimum the limit will probably result in a 1,000/1, errors rejected/valid codes rejected ratio, which seems acceptable to me.

For any errors displayed, your best course of action is to convert the code to the RAW format the device lends itself to and check out what it's doing.

Edit: Minor change to correct the single-line fill count added. Re-download the file if you downloaded it before noon EDT on July 22nd.

[Kry_out_lowd]

hey man, can u make one that converts saves?...i really need a program that can convert max drive saves to xport and sharkport, etc. and vice versa

[Pyriel]

Late last night I was looking for a way to screen out two-line directory codes and the occasional master code that contains a 2nd verifier line and I stumbled upon the method used for creating multiple directory levels in the code list. I made some codes to test it out if anybody would care to take a look. The codes are for La Pucelle: Tactics, but that's just a detail.

CHAR 1 CODES FOLDER
02EE4000 00500800

ITEM ABILITIES CODES FOLDER
02EE4000 10480000
D29EE62A 4457657D

ABILITY 1 CODES FOLDER
02EE4000 20480002
D29EC62A 4457657D

CLOSE CALL
02EE4000 80480005
0032EE0A 00000001

CHAR 1 CODES FOLDER
9YYU-1JP0-HXM8C

ITEM ABILITIES CODES FOLDER
3G9Q-NTKH-64ZTW
JB9E-91K0-74840

ABILITY 1 CODES FOLDER
2UKU-YGY8-VYAE4
8UQH-1851-EZQPC

CLOSE CALL
AJTE-8PA7-AN2JK
38QY-7MDA-GKEPN

Creating the sub-folders involves carrying over the expansion data to subsequent lines. I'm still not entirely sure how the size parameter in the verifier affects the expansion data. When I thought I had it figured out, I found a code that defied my reasoning. It could just be an error on CJ's part. At any rate, it's possible to use a relatively simple set of rules to create the directories. I'll write it up later if there's any interest.

[GMO]

interesting... i'll looking into this right now

[nojunkmail4me]

Link to file is now invalid

Anyone know where a good link to maxconvert is?

[Azagthoth0677]

http://www.codemasters-project.net/p.../request.php?6

[raijinsky]

Bump.

I'm looking for updated converter programs. Can't seem to find any that can convert CBV7 codes to GSV4. Particularly, looking for MS Saga: A New Dawn codes, and freaking Gameshark isn't updating.

If anybody can, couldja convert these to gamesharkV4?

master code 90133094 0C09B3B7

infinite money 1B39E900 0000FFFF

max money 2BCBDF2D 0098967F

1exp for level 50 4AA462F1 C4FA01F5
0A22080D 00000000