8 Apr
2013
8 Apr
'13
7 p.m.
On 08/04/13 12:07 PM, Chuck Guzis wrote:
On 04/07/2013 03:20 PM, Toby Thain wrote:
Bitsavers has an 11/40 system manual which lists some timings for Unibus
core:
http://bitsavers.trailing-edge.com/pdf/dec/pdp11/1140/1140_SystemManual.pdf
(4th printing, January 1975)
"MF11-U/UP Memory is a read/write, random access coincident current,
magnetic core type with a maximum cycle time of 980ns and a maximum
access time of 425ns."
However, there is no MOS memory listed for comparison.
According to its system manual, the 11/34 (of which I own a couple)
supports either core (MM11-CP or MM11-DP) or MOS (MS11-EP, MS11-FP,
MS11-JP).
I found the MM11-E core memory manual:
http://bitsavers.trailing-edge.com/pdf/dec/pdp11/memory/MM11-E_CoreMemoryMa…
"cycle time is 1.2?s and access time is 500ns"
According to the MS11-M MOS memory user guide,
http://bitsavers.trailing-edge.com/pdf/dec/pdp11/1144/EK-MS11M-UG-001_Users…
(May 1979)
DATI/DATIP (Memory) Typ. Access time 490ns; Typ. Cycle time 450ns
A later document, MS11-P MOS Memory Technical Manual, Oct 1982, gives
490ns and 680ns for typical access and cycle times (DATI).
Not very conclusive... It seems that some core models were slightly
faster than some MOS models and vice versa.
--Toby
That's odd, I seem to recall my handbooks
claiming lower latencies for
(Unibus?) core than the MOS products in the same range, but don't have
the right handbook to, uh, hand.
I don't know from UNIBUS, but the CDC 7600 used 270 nsec core, which, in
1969 was much faster than MOS. ...
The "exchange jump" on the upper CDC systems (6000 and 7000) used a
read-modify-write cycle to swap the contents of the registers with an
area in memory. It was the fastest way to save and reload the enire
register set (using non-privileged instructions was far more
time-consuming, due to an odd trick used to save and restore the first
register of a set--often used as a quiz to new programmers).
--Chuck