< Options under consideration:
<
< o Sockets for two 1822/2101 256x4 SRAMs, the same style used by
< the original Elf.
one 2116 outweighs that.
Yes, but I was considering allowing the builder to have a more traditional
design, if they so desired. The downside is routing complexity and loss
of real estate. 44 extra holes isn't a major cost factor, in and of itself.
<o A 1851 programmable input/output port -or- a
pair of 4508 latches
Use the one that is available.
Both are available. Harris makes the 1851 programmable I/O. It takes three
N-style I/O ports, and up to two EF inputs or the INT. The 1851 has the
advantage over the 4508's of allowing bit programmable I/O and strobes,
as well as just a byte in and out. The 4508 takes up a little more
real estate (2x24 pins vs. 1x40 pin), but costs *lots* less (less than $2
for the pair, vs. $10 for the 1851).
The VIP used 4508s. The 1851 is cool. I was throwing out both for
consideration.
< wired as one input and one output port (as in
the COSMAC VIP)
Handy and useful, do the ports.
There will be ports. The question is how to implement them.
< o A 1854 UART
Desirable.
Indeed. I have had an 1854 for years. I have always wanted to be able
to use it.
< o Either a dual 4042 address latch or a single
4508 latch. One or
< the other is needed to implement more than 256 bytes of memory.
< This would be optional in a machine built with dual-256x4 SRAMs.
Consider 74hct373.
I'm attempting to preserve some amount of consistency. Yes, HCT is CMOS,
but it's as jarringly new as a 62256 compared with the 1802. Given that
two 4042's are compact and cheap (the 4508 is a smidge cheaper than two
quad flops), I don't mind staying with an older part.
< o A pre-programmed 27256 with a
"library" of simple programs
< (Accessible by enabling the ROM switch, setting the program
< number in the switch register, then setting the control switches
< to "run").
Good idea! I have sources for minimonitor and UT4(RCA) plus a buch of
others in hex from articles.
Cool. Submissions welcome. One catch. The implementation of this is
to put INP 4, PHI 0 as the first two bytes in the ROM. This will read
the switches, and jump to $XX02, presumably where $XX is > $80. As long
as the ROM code doesn't need *every byte* in the first page, we can do this
easily.
So far, there hasn't been a groundswell for 1822/2101 support. A few more
negative comments and I'll turn over the real estate to I/O.
-ethan