Re: LZ: Extended mem. format

Subject: Re: LZ: Extended mem. format
From: Mel Tsai <tsaimelv@pilot.msu.edu>
Date: Sat, 21 Sep 1996 14:11:45 GMT
In-Reply-To: <2D49EAA8.61B06980@foma.wsc.mass.edu>
On Fri, 28 Jan 1994 23:48:40 -0500, you wrote:


>No. Here are two practical solutions to extra memory for the calc...
>
>Passive. We interface a microwire eeprom (its three wires, clock, data
>in, and data out. I believe its possible to combine the data in and data
>out using passive components, but well have to look into that). I Have a
>500k-byte EEPROM with a microwire interface, if that isnt enough, they
>have a 1-meg one you can buy. Basically we wire this to the link port,
>and do the rest in software.
>


I've completed a circuit that will interface the Microwire Chip.  It
isn't very easy to multiplex the data in and data out lines, but I
think I've done it using about 3 other 74LSxx parts.  All my findings
will be posted on my web page some time tonight or tomorrow (I'll send
the address to the list).  The big headache is that the chips only
come in surface mount or PLCC packaging, which will make soldering a
big PAIN!!!


>Active. I like this one better, because you could add a lot more to it.
>Basically, you take another microprocessor (Probably a Z80, since we
>already know how to program it) We could hook it up so that if your calc
>got reset, you could program the thingy to automatically send a backup
>to the calc if you did an input "cblget",blah or whatever on a cbl calc
>and go to link receive or whatever. This way there could be variable
>send/raw send/etc. The only problem is that i doubt if anyone has had
>any experience working with microprocessors and stuff (Me Included, i
>got the idea from my Z80 book). Wed at least need a microprocessor, rom
>code, actually, would probably be eeprom that would be locked by a
>special switch so that updates could be done without needing a burner.
>Wed also need either static ram or eeprom, or both maybe. We could leave
>the scematics open-ended (same with the rom code) so that if someone
>wanted to use eeprom for no-power storage, and static ram for fast
>access, although it would be through the link port, so it would be
>pretty slow. Anyway, i think the practically of this is limited by the
>fact that it would be to big and bulky.
>


I have enough experience with microprocessors, but I can say right now
that it's not worth it.  For one thing, finding Z80 processors is
difficult to do these days.  I think NTE makes Z80 clones right now,
but they probably charge an arm and a leg!  Out of my 5-10 electronics
supplier catalogs, I have yet to see the Z80 listed.  That doesn't
matter, however, because the Z80 wouldn't be the best choice.  If you
absolutely must use a dedicated microprocessor, use some of the single
chip microcontroller's like the 68HC11, the Z8, the Basic Stamp, or
even the 8051.  The Z8 would be nice because it has the Z80 core.  In
any case (like I said in the expander doc's) using a microcontroller
is just not worth the effort.  Most of the uP's functions could be
emulated with passive components.


The best solution (as you'll see soon) is the National Semiconductor
Serial Flash chips.  All the info will be posted to my web page
tonight.
References:
From: