I have found a nice CRC web site that allows one to change many of the settings of the CRC to try out various polynomials. http://www.gpfn.sk.ca/~rhg/csc8550s02/crc.html I've been tossing in a few numbers at random to see if I can see what variation of the CRC-16 polynomial it is. I believe that Fluke used the standard CRC-16 formula with a preset CRC value as there is mention in the docs about dividing the data into a fixed number. Now to figure out the number... I'm just using an old posting of mine where I tried some two byte data chunks on the Fluke ROM Signature generator...but I'm not getting very far yet. I don't know how to solve hexadecimal polynomial problems...considering that we pretty much know the formula X16 + X15 + X2 + 1, and the data and the result, but the mystery "binary number" is the tricky bit ... I'm way too rusty (like 30 years since I last did this) to figure this one out without a bit of help... Sig. - DATA (two bytes) 93D0 - FF 00 9351 - FF 01 9390 - FF 02 9311 - FF 03 ..... 93AF - FF FE 932E - FF FF John :-#)# At 10:45 PM 12/10/2002 -0700, John Robertson wrote:
Seed file? I used the Pacman 6E as I recall. Just ran the calculator utility included with Hexedit (http://www.expertcomsoft.com/) a nice windows file editor - search and replace etc...
There was mention that the signature was created using "The ROM signature is obtained by successively dividing the data in ROM by a binary number" in the manual as you may recall...what order would the data be divided? Would this matter?
John :-#)#
At 11:38 PM 12/10/2002 -0400, Kev wrote:
Ah yes but what are you using for a seed file for the CRC 16 /32 calculators?
I belive the Fluke version is possibly based on some of the original articles that were in BYTE circa 1983. I'll need to research that a bit more.
Kev
Still looking for this one - what is the algorithm used to generate the ROM Signature in the 9100/9010s? Have tried almost all the CRC programs I could download, none were correct. It's not CRC-16, CRC-32, CCITT, CHECKSUM....
Anyone get any further?
Attached is the file that I used to try and identify the signature and these are the results:
CRC-16 =58B8 CRC-32 =C1E6AB10 CCITT =E63C
(the "Correct" Fluke signature is = 8B1A)
The file is from a video game, it's a 4K size chunk of the main program...
John :-#)#