From Game Developer Research Institute
Revision as of 05:35, 17 May 2012 by Andlabs (Talk | contribs)

Jump to: navigation, search

Disassembling random games to take a peek is fun... I was recently reminded of a similarity I found in a number of games that used a similar system for their main loop routines, involving arrays of data including a function pointer that get loaded in a special way in RAM and then run. I did a binary search for a portion of the common code and came up with this:

Jantei Monogatari (Japan).md: 00008930
Metal Fangs (Japan).md: 00000378
Ninja Burai Densetsu (Japan).md: 000029be
Rent a Hero (Japan).md: 0000121c
Sword of Vermilion (USA, Europe).md: 000010ac
Vermilion (Japan).md: 00001018
Wrestle War (Japan) (Beta).md: 00000842
Wrestle War (Japan, Europe).md: 00000846

I don't know if you noticed this already in a code comparison but eh. Metal Fangs's credits are initials only and are plaintext in the ROM and I am looking for Jantei Monogatari's. I'm not sure if this says anything about GDRI-010; I'm putting it here because of Wrestle War's confirmation - Andlabs 10:51, 30 March 2012 (CDT)

Well it doesn't have credits (apparently) :/ (set FFFFD0:0053 at startup to replace the intro cutscene with the ending cutscene) - Andlabs 11:00, 14 April 2012 (CDT)

Dynamite Dux (arcade) does the same thing! That game does credit a programmer "duke" as Assistant Programmer... is it our Soul Duke? (If you deinterleaved the ddux1 ROMs, which aren't encrypted, the code I looked for above starts at $4D2). The data structures and values aren't exactly the same, of course. While the screen mode function is the same, after that I can't see many code similarities... so I don't know if the programmer's style evoled or I'm not looking hard enough or this guy didn't work with the same people every time - Andlabs 00:35, 17 May 2012 (CDT)