|Re: What is this "ADV HEPAX module for the HP-41?|
Message #5 Posted by Dan Grelinger on 17 June 1999, 8:43 p.m.,
in response to message #4 by Chris Ctotti
This is truly an amazing device. If the inventor of this piece is still around, I would like to congratulate him on his great work. Alas, since there does not seem to be many of these around, he must not have sold many.
For the auto ROM addressing to work properly, the ROM must default at turn on to an address that is definitely not used. This would either be page 4 (reserved for diagnostic, and executed only at machine turn on), or in the address corresponding to the port the module was plugged into. I have a way of checking this.....
Wow, I did not know that ENBANK3 (140) and ENBANK4 (1C0) instructions existed. This means that 32K ROM modules were possible in each port of address space. A programming nightmare, but possible. Theoretically, a 128K ROM (or RAM) module could be made that logically occupied all 32K of port addressed ROM space, and banked switched 4 times as much memory using the bank enable functions.
I have confirmed that the memory is lost immediately if the module is removed. One interesting feature of this module is that it stores programs and data very similarly to the Extended Functions Module, with one major difference. When a program is saved, it automatically becomes a Catalog 2 program, executable. This is something that is not done in the EFM. Another major difference is that when used as data registers, the 16K of ram converts to 2610 registers, a healthy number.
Thanks for the information Chris, you are a scholar and a gentleman. (Assuming that Chris is used in the masculine sense, of course).
By the way, is the HPExchange article you mention on either the Museum CD, or another one?