A couple of DRAM memory stick questions ??

[Bill Davidsen]

Patrick O'Callaghan wrote:
> On Wed, 2009-09-30 at 10:55 -0400, William Case wrote:
>> I'll take that information to the bank.  To state it another way just
>> to make sure I've got it.  A typical physical address goes to, or
>> points to,  8 + 8 + 8 + 8 + 8 + 8 + 8 + 8 cells arranged side-by-side
>> in a line on an individual DIMM/DRAM stick. 
> 
> +8 more if it's parity memory, or more for ECC memory.
> 
>> I suspect that by thinking of address as divided into bytes rather
>> than a single 64 bit word (dword, qword, -- pick your author) there is
>> a natural division for instructions, numbers and characters within the
>> 'word'. Or, is there some physical reason why it is thought of as 8 +
>> 8 ...
> 
> It's the data that's stored in units of 8 bits. When addresses are
> stored then of course the same applies. When they're on the address
> lines of the memory bus, they may be in groups of 16 or 32 or 64
> (depends on the bus design). None of this matters to you as a
> programmer.
> 
> Note that the pedantic name for a group of 8 bits is "octet". A "byte"
> is the number of bits required to represent a character in some
> encoding. Nearly all modern machines have 8-bit bytes and are
> byte-addressed, so we tend to equate "byte" with "8 bits", but I've used
> machines where addresses refer to 12-bit words, and the DEC-10 famously
> had 36-bit words and a configurable byte size, usually set to 6.
> 
The GE-600/6000 line had 36 bits also, and could use 6 or 9 bit characters. We 
used ASCII with the unused high bit as end of string. MULTICS was developed on a 
GE-645, leading to UNIX, leading to Linux. Sort of.

Nomenclature: 8 bits is a byte, 4 bits is a nybble, 2 bits is a quarter. That's 
probably only funny to old programmers in the USA, sorry.

-- 
Bill Davidsen <davidsen at tmr.com>
   "We have more to fear from the bungling of the incompetent than from
the machinations of the wicked."  - from Slashdot