Re: TI-H: Expander, etc.

To: ti-hardware@lists.ticalc.org
Subject: Re: TI-H: Expander, etc.
From: Bryan Rittmeyer <bryanr@flash.net>
Date: Fri, 22 Aug 1997 10:03:42 -0400
References: <8DE13EC.000D0005D6.uuout@min.net>
Reply-To: ti-hardware@lists.ticalc.org

JUSTIN BOSCH wrote:

> I was just thinking about something: the expander and the extender (not
> sure about this one) use AA batteries to power them.  Why not just use
> regular memory and a lithium battery like the calc itself?  You could
> write to it infinite times and it would not "wear out".  One lithium is
> not much more than a set of AAs.

Because A) Normal SRAM (DRAM requires massive amounts of power to
refresh)
costs quite a bit more than the serial flash EEPROM the EuP and the ESF
use(d), B) RAM requires continuous power, which means that your
batteries
would die MUCH sooner than if the memory was non-volatile (I.E. EEPROM)
and C) Traditional memories require a large amount of I/O, which we do
not have.

And, no, The EuP and the ESF also (correct me if I am wrong, Mel) can/do
use 9Vs.

In actuality, the Atmel DataFlash chip that I am using in the EuP is
much
better in endurance than the NSC chip Mel Tsai used in the ESF. With the
new fab technologies, EEPROM memories are getting closer to the
reliability
of traditional RAMs. However, the Atmel DataFlash has a price for its
extreme durability (I.E. it will essentially never wear out, even under
heavy usage) -- The entire array must be rewritten after every 10,000
write cycles. With the EuP this will only take ~4 minutes so its no
big deal - Think of it as defragging.

Bryan Rittmeyer
bryanr@flash.net

References:

TI-H: Expander, etc.

From: justin.bosch%tfexchng@min.net (JUSTIN BOSCH)