Subject: [stella] Clones From: Thomas Jentzsch <tjentzsch@xxxxxx> Date: Sat, 7 Oct 2000 18:44:24 +0100 |
Hi, i was always wondering, how many real different games where programmed for the 2600. When i downloaded a file with more than 1400 different roms, i decided to find out. I wanted to let my PC help me. So i wrote a little Borland Pascal program which compares rom-dumps. This is done by a pseudo dictionary-based compression, which finds duplicate code or data fields. Each rom is compared against all other ones, and when i shows some relationship to another rom, the later is also compared against the first. This gives two compare values, which are written into a text file. Before i started the test, i removed all multipack-dumps, test and simple demo roms and finally counted 1075 rom-dumps. The result is in the attached CloneSpy.txt-file. The ListSpy-program shows the txt-file as lot of matrixes where similar roms are grouped together. The coloured (red..yellow..green..blue) numbers show how similar two roms are. All rom-names were created by good2600 v0.9994. Though these tests are far from being perfect, they should help to get some answers about the history of those roms: - which original rom was used for a clone/hack? - were has code been reused for different games (Brick-Kick and Pac Man, Mythicon)? - how big are the differences between PAL and NTSC versions? - how different are versions of the same game (Dodge 'em)? - are some suspected games really clones? - does an oversized dump also contain another rom (Combat - Tank Plus)? Most informations are only statistical, but they give a hint where to start (disassemble). What do you think about it?
Attachment:
CloneSpy.ZIP
Description: Zip compressed data
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: Re[2]: [stella] I did it!(SFCav, Mark De Smet | Thread | Re: [stella] Clones, Piero Cavina |
Re: Re[2]: [stella] I did it!(SFCav, Mark De Smet | Date | [stella] I haven't received any mes, Joe Dowlen |
Month |