Monday, June 26, 2006
Base+Offset notation (or why we start counting with zero)
Every now and again, I get the question about why we starting counting things such as arrays, offsets, etc. with zero (0) and not one (1). The answer is simple, when specifying a data structure, we normally specify the byte (or whatever unit) offset for the start of a field for a specific data structure. Think how a computer would get to a specific element of a list. Let's say we want the third element in the list (the offset is 2). The start of the list is the first element. We move to the next element, which is at offset 1, and then we move to the next element (the third item) which is offset 2. We now read the contents of that element. When we start counting with offsets, the first byte is zero bytes from the start (since it's the first byte). Here is what it looks like graphically:
For example, the FAT boot sector starts out like this: (Note: this was taken from http://support.microsoft.com/kb/q140418/)
0 1 2 3 4 5 6Let's say we want to get to element C (the third element, offset 2). If "A" is the start, then it is zero bytes from the start. That means, that A's offset is 0. There is a caret (^) beneath "A" to show the current position. Next we move forward one element (offset = 1) to the second element, which looks like:
A B C D E F G
^ (offset from A = 0)
0 1 2 3 4 5 6Finally, we move forward one more element (offset = 2) to the third element, which looks like:
A B C D E F G
^ (offset from A = 1)
0 1 2 3 4 5 6
A B C D E F G
^ (offset from A = 2)
For example, the FAT boot sector starts out like this: (Note: this was taken from http://support.microsoft.com/kb/q140418/)
Field Offset LengthNotice that the "Bytes per Sector" field (which tells us the size of a sector, usually 512) is at offset 11. This means that this field starts at offset 11 (a.k.a. 11 bytes from the start, the 12th byte, etc.) where the offset of the first byte is 0 (the start).
----- ------ ------
Bytes Per Sector 11 2
Sectors Per Cluster 13 1
Reserved Sectors 14 2
FATs 16 1
Root Entries 17 2
Small Sectors 19 2
Media Descriptor 21 1
Sectors Per FAT 22 2
Sectors Per Track 24 2
Heads 26 2
Hidden Sectors 28 4
Large Sectors 32 4
