Casio CZ/ VZ/ FZ - Pro Series

I suspect that the VZ doesn't use software to detect the difference 
between a RAM and a ROM card. There are other conditions that must be 
met other than just the WE pin, before writing can commence. 

It may be that the VZ looks for a completed circuit between the CE, 
CE1 and RA19 pins on the card before writing, all of which connect to 
the NOR gate chip and then go to the CS pin on the RAM chip, but are 
all not connected in the ROM chip. 

Also, CE1 on the RAM chip is permanently high, but not connected on 
the ROM chip. Without all those conditions being met, the VZ perhaps 
will not go into transmit mode. The "write fail" message may appear 
as a result of a negative scan result of CE, CE1  and RA19 lines, not 
because of failed data transmissions. Maybe. 

Having said that, if none of that is true, and the VZ  is able to 
blindly "write" to a ROM, the 1k resistors will indeed be protection 
against any data signal, I would think. But I think the above 
scenario is the case, rather than this one. That being true, IF it is 
true, then the extra data line RA15 may be availiable for a bigger 
RAM. In any case, I will be trying this out as soon as I get my VZ10M 
in about 6 weeks. In the meantime, I'm going to breadboard up a 
prototype card. I've already located a source for the standard RAM 
chip:

http://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?
langId=-1&storeId=10001&catalogId=10001&productId=279857&pa=279857PS

And a 16x one here:

http://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?
langId=-1&storeId=10001&catalogId=10001&productId=280049&pa=280049PS

Both very cheap.

Here's the NOR gate:

http://www1.jaycar.com.au/productView.asp?
ID=ZC4827&CATID=33&keywords=&SPECIAL=&form=CAT&ProdCodeOnly=&Keyword1=
&Keyword2=&pageNumber=&priceMin=&priceMax=&SUBCATID=884

Cheers, Graham




--- In CZsynth@yahoogroups.com, "Scott Nordlund" <gsn10@...> wrote:
>
> >In any case, I don't think the VZ sees any difference between a ROM
> >and a RAM, except that it will fail to write to a ROM because the
> >ROM's write line is not connected (and wouldn't work if it was,
> >anyway.). This being the case, using a larger RAM  chip will give 
you
> >an extra bank WITHOUT DIP switch selectors for extra address lines,
> >because you can use RA15, which is disconnected on the RAM card, 
but
> >is utilised on the ROM, being 2x larger than the RAM. For even more
> >banks, you will need DIP switch selectors though.
> 
> This may be the case, and I notice on mine that it does appear to 
try to 
> write to the ROM card before giving an error message.  But I was 
thinking 
> that this might not be a great idea- the ROM chip doesn't have the 
WE pin, 
> and what are I/O pins on a RAM chip are just outputs.  I would 
assume that 
> the VZ just blindly tries to write the data by asserting WE and 
then sending 
> the data (if it hasn't got any way to tell), but on a ROM chip this 
results 
> in output connected to output, basically a short circuit.  There 
are 1K 
> resistors on the I/O lines, but it still seems like it's not the 
best way to 
> do things, and I think any decently-engineered product wouldn't do 
that.
> 
> But regardless.. how, then, would it give an error message after 
trying to 
> write to a ROM card?  If the VZ doesn't differentiate between RAM 
and ROM 
> cards, there's no way to reliably detect a write error.  Even 
comparing the 
> data on the card with the internal VZ RAM after the write isn't a 
foolproof 
> way to do it.
> 
> However, now that I think about it a little more, there are ways to 
detect 
> the difference in software.  If there's some unused portion of the 
memory 
> space, it's easy to tell if the highest address bit actually does 
something: 
> If ROM card address 0x0000 (bank 1) contains the value "10101010" 
and 
> address 0x4000 (bank 2) contains "01010101", 0x0000 XOR 0x4000 
= "11111111". 
>   On a RAM card, it would simply refer to the same byte, so 0x0000 
XOR 
> 0x4000 = "00000000" (regardless of what data is present).  Whether 
the VZ 
> does it on start up or just before attempting to write is a 
different issue, 
> but it would thwart the 2-bank RAM card idea.
> 
> At any rate, if a RAM card indicates CARD (1) and CARD (2), this 
would 
> confirm that the highest address bit is being used, just not 
connected to 
> anything, and that RAM and ROM cards are indeed detected the same, 
at least 
> on startup (and thus a 2-bank RAM card can probably be used, unless 
it 
> detects before attempting to write).
> 
> _________________________________________________________________
> Get a preview of Live Earth, the hottest event this summer - only 
on MSN 
> http://liveearth.msn.com?source=msntaglineliveearthhm
>