Talk:X86 Assembly/16, 32, and 64 Bits

Why not use offset addressing in 32bit mode? Then instead of having access to only 4 Gigs of memory, you'd have access to a maximum of roughly 68 gigs. FFFFFFFF:FFFFFFFF = FFFFFFFF * 10 + FFFFFFFF = FFFFFFFF0 + FFFFFFFF = 10 FFFF FFEF Nikayah (talk) 03:29, 2 July 2009 (UTC)

Mixup of different terms in section "The A20 Gate Saga"
In section "The A20 Gate Saga" it is stated you say first that So far, so good; it is consistent in itself, but not quite right.
 * FFFFF is equal to 1MB
 * FFF0 is equal to 64KB minus 16 bytes.

FFFFF is the address of the most significant byte in a 1MB region of memory, but the size of a 1MB region is usually denoted by 100000; otherwise we had a discrepancy in describing 1 byte alone, with the address 0 and size of its region... umm... 0??? You know what I mean.

You do (know) because just three paragraphs later you make use of the notation, that I just described, yourself, when saying that a reference to 1MB would get wrapped to address 0 etc.

Though this has no effective implication to todays programming practices, I don't think it's like nitpicking :-) Would be nice to see this fixed. Thanks.

Furthermore, in the line just above the first claim, where you calculate those hex values, none of them (as in this post) is suffixed by the 'h' character, but the 10, which was just above declared to be in hex already...

CU Jagged