Re: TI-H: Parallel Memory expansion


[Prev][Next][Index][Thread]

Re: TI-H: Parallel Memory expansion



Mel Tsai wrote:
> 
> >> The Expander II and III are both in development right now, and should
> >> be available within a few months.
> >>
> >Cool; I hope whatever I build will have it's own niche though. :)
> 
> I'm not saying that you won't do it, but *guess* how many people have
> emailed me describing their own expander project that will use
> parallel flash chips or some other form of mem?  You're #6!  Many of
> them are already far on their way.  But frankly I don't see one coming
> out, at least not anytime soon.
> 
The real key is forming a stable interface (zshell) to the calculator.
My inital schematics are really done, for the most part. I'll post
something on a page soon.
> >
> >The schmatics I've seen need 3 wires, a Clock, data, and the ground,
> >which the TI-85 can do just fine.
> 
> Are you talking about I2C parallel/serial I/O converters or actual
> serial flash EEPROMS?  There is no 2 wire interface high capacity
> serial EERPOM on the market today, that's a fact.  One *is* coming out
> though but I probably won't be using it unless it's very cheap (I
> haven't see pricing specs for it yet).
> 
Well, (Not including ground,) I2C uses 2 wires, so it works well, but
from what I have seen in datasheets (I have not recieved one yet to test
this), There are some Serial FLASH memorys that should be able to use
only 2 wires also. When my orders finally show up, I'll know for sure.
> 
> Microcontroller's have TONS of advantages here.  First of all, they're
> just one part compared to the many needed to interface to a serial or
> parallel chip without one.  Second, they are easy to program once you
> have the equipment.  Third, they are the cheapest alternative and
> require the least amount of board space (you can get a relatively
> powerful 16 pin microcontroller that could do the job for about $1.25
> qty 1).  Fourth, they allow you to do FAR more important things such
> as backup transferring, built-in file transferring intelligence,
> better overwrite security, power management, etc.
> 
> What do you mean by "unable to address memory"?

I meant that  a PIC microcontroller is unable to address external
parrallel memory, but as it appers your using the PIC to interface to
one of the new serial chips, it doesn't matter.

> 
> The controller I plan on using for the Expander III *is* more powerful
> and tons faster than the Z80 inside the TI-85, and it's also much
> faster than the 68k in the 92.  But that's not a bad thing!  Wait to
> you see what this baby can do and you'll become a believer :).

> 
> Of course they can, but the disadvantages of not using a
> microcontroller FAR outweigh the advantages, as anyone who's
> experienced with them will tell you.

I must admit that I have little experience with PIC microcontrollers or
similar devices, and that's why I shied away from them origionally.

'Course, the ideal way to access the Z80 in the calculator would be to
pass a ribbon cable out the side containing the 8 lower address bits and
the data bus, and the read + port bits also. 'course, that would equate
to about 20 wires. But, if someone was going to cut a hole in the side
anyways. (Like for the SPinTerface) Direct bus access would be
infinitely useful.

If someone REALLY wants to get the full use out of a Microcontroller or
similar device, it would be interesting to see if you could interface to
the PCMCIA standard.

Anyway, just my thoughts.

Charlie Goetzman
tgoetzma@lax.net


Follow-Ups: References: