25 Aug
2012
25 Aug
'12
3:59 p.m.
On 08/25/2012 05:14 PM, Tony Duell wrote:
That would be insane as 8041 was a 1Kb (code space) embedded micro with
and trivial IO port designed as a intelligent IO device. Boot code there
would have to be small and the main system would still need a boot to get
the code out as the 8041 could not be a bus master
(it could be a DMA target/source).
One can be creative but that seems like the hard way to get things done and
every system I've seen that has the 8041 used all the code space to do more
useful things (printers, keyboard scanning or IO for examples).
Allison
On 24 Aug 2012
at 19:44, John Elliott wrote:
That is what I was told
by one of the designers at Amstrad.
Using the 8041 as a boot ROM would be problematic IMHO. There could be
code i nthere to preset the boot ROM bytes one at a time on a noraml read
cycle I suppose -- i nfact that could eb why the PROG pin is conencted to
a signal o nteh SIC alled NBOOT. Readign the ROM directly involves
overvoltaging one of the 8041 pins, I can't eee the circuitry to do that. In theory you could replace the boot ROM
(desolder the 8041, read
its
contents, and burn an 8741 with the new boot image. Or even an 8742,
if you wanted more than 256 bytes of boot program). In the 9512 it
would be even easier, because the corresponding microcontroller used
an external ROM.
Hmmm. Tony seems to think that the boot code resided in the
ASIC,
but that would have been the difficult way to go about it. Waht does a PCW do (or not do) if powered up with the
pritner
microcontroller missing.
Thanks for the info about the hard disk providing
its own boot code.
That's definitely a starting point.
Wht does the MDIS/ signal on the expansion
connetor do? It gors to the
gate array and nowhere else/ From the name it might be 'Memory Disable'
and might, indded, disable the internal boot ROM if pulled low at the
right times.
I didn't relaise before I looked at the schemtaics that the video and
sync signals are on thet expansion conenctor too.
-tony