Mario Michael da Costa wrote: > Never saw my name on a website before, done my ego a whole lotta good, > thanks.
You're welcome! > Were the sizes of the 2 iso images different ? because that is the > only thing that could the message digest. Before computing the MD5 > digest, the message (iso image) is padded till it's length is 64 bits > less than a multiple of 512 bits. so if the 2 sizes were different, it > is possible that they would be padded differently. heck if they were > different, the MD5 digest would be different in any case. Don't know about the sizes, and they are now deleted. If I get a chance to experiment again, I'll check that. Thanks for the explanation on padding -- I was afraid that different isos (from different software) might be padded differently -- I'll add this to the webpage. (We may still not be positive, but this sounds like a straightforward "algorithm".) > As part of my *research* on MD5, i downloaded the c source code of > md5. It compiles on linux quite easily. The code though was a bit too > geeky for me to comprehand, the algorithm is pretty simple. I can mail > you the c source code offlist, if you want to try and compile it on > windblows. Yeah, go ahead and mail me the code -- don't know when I'd have a chance to look at it, or what I'll be able to pick up, but learning C and C++ is on my ToDo list. > > Over the coming weekend i'll try experimenting on my own. Great, send me any results you get, or add them to the Wikilearn yourself. regards, Randy Kramer Reminding me (and all of us) of the link: http://twiki.org/cgi-bin/view/Wikilearn/Md5sumsAfterBurning
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
