14 Jul
2015
14 Jul
'15
8:18 p.m.
On 07/14/2015 09:42 PM, ben wrote:
I guessing ( no schematic handy) that they made the 360
register file
easy to decode and build with latches.
Not just the register file, but the entire machine. So, all
the hidden registers in the RTL description,
such as storage address register, storage data register, the
temp registers that held the data from
a selected local store register, etc. were ALL latches.
By the way, the 360/30 through the 360/50 did not have an
electronic register file (local store in
IBM terminology). On the 360/30, they were implemented in a
separately-addressed extension to
main storage. On the 360/40 and /50, they were in a small,
fast core memory unit. So, on all these machines, the
register set could only be accessed one location at a time.
I don't know on the 360/65 whether they had dual ports to
read the electronic registers two locations at a time, but I
might guess not.
yes, the 11/45 was implemented in pretty early TTL, and the
original 74xx family did have some 6-wide latch devices that
used less power than the equivalent 6-wide full D FF.
Jon
The 11/45 used
TTL ICs, so real FFs were available in
that technology,
although they may have used latches as well.
I have seen some 11/?? schematics on bitsavers that the
alu uses AOI gates
and includes a latch term.