Re: Restoring a NorthStar Horizon, problems with SRAM board

Hi,
A quick update on the progress on restoring the NorthStar Horizon. I
wrote another program, a terminal loop back to repeat whatever is sent
over the serial port, and it worked well. Also got the HDW80 monitor
working by just changing the ORG to $F800. The 2708 simulator based
on the DS1220 and shim socket circuit seems to be working fine.

Glad to hear it - btw, one thing to watch for when using a Dallas chip as
an Eprom emulator - the Dallas has an internal power monitor and it may
not "come up" as quickly on power-on reset as the CPU - the symptom
is that the system doesn't come up with power-on, but works OK when you
reset it afterward.

Already the HDW80 monitor is helping me find some very odd behavior
and I am starting to suspect the FDC board is bad. The NorthStar MDS-
A-D FDC is a memory mapped IO device so I would expect some behavior
when reading memory in its valid range ($E800-$EBFF) as far as I
know. However, whenever I read memory in $E000 or $F000 range outside
of the FDC memory map, I am seeing floppy drive activity like drive
lights turning on, motors starting, etc. Either the FDC memory decode
logic is messed up or the CPU board is generating junk addresses on
the bus.

It does sound like iyou have a decoding problem. Can you read the MDS
boot PROM at E800? If so, try removing your RAM card, and see if the
MDS ROM shows up in the places where it shouldn't.... (If you want to
knoe the content of the MDS ROM, it's contained in my Horizon simulator
which is available from my site).

The monitor (which is called HDM80 btw - Hardware Debug Monitor) has
loop read/write functions - which puts the system into a two instruction
loop reading or writing the specified location - this is very handy to
diagnose decoding problems - IIRC you obtained a scope? - if so,
trigger on the RD (or WR) signal, and follow the address lines see if
they are correct.

Note that the NorthStar boards being socketed are prone to corrosion
and crud causing bad conenctions - I've had to remove every IC and
clean the pins/sockets several times over the years - Look for visible
corrosion, discolored pins etc.

I am leaning against the CPU board generating junk addresses since I
can reliably read the FDC PROM and the boot ROM memory locations and
if the CPU addresses were junk, I'd think I would be reading trash
values or at least unreliably. I can read and write reliably to the
SRAM on the Vector Graphics Flashwriter video card located at $F000
(another neat discovery). In addition, I am reliably communicating
over the serial port which would be impacted by bad CPU addresses as
well since its program is at $F800 and IO ports are at $00 range

Ah - so you've already tried the FDC PROM - as I noted, remove the RAM
card (and any other memory mapped cards) and see if it appears
anywhere else.

I'd also lean away from bad addressing by the CPU - to activate the
drive, you need to be within the page EAxx (or is it EBxx...), which is
a fairly specific address to have keep coming up - It's more likely that
the FDC is not decoding (if you've gotten an open connection due to
IC/socket corrosion, it could be assuming a select that isn't there).
Use loop read/write, trigger on the appropriate signal, and follow
the addressing through the bus and decoding logic in the FDC - An
S-100 extender card is very handy...

Later today, I plan on more investigation with HDW80, the SRAM memory
board and the Flashwriter. The first test is to see if the SRAM board
is working properly. If so, I may move up to the other monitor for
more general debug utility. I would also like to know if the video
board can correctly generate video by just writing directly to the
video SRAM.

If HDM80 appears to read/write the RAM correctly, take the memory
test from MON85 (MON85 is about 4K and won't fit in the 1K ROM
slot unless you strip away some functions) - this does a reasonably
comprehensive RAM diagnostic.


PS, I tried to find the documentation for the ASM85 program but
couldn't. Is it online somewhere? I would like to create a listfile
when the ASM85 assembles a program. I review those before running to
try to catch errors before burning ROMs, etc. Thanks!

It's part of my XASM package, which is available from my web site. If
you install it as a demo, you will have access to the full documentation,
however in brief, the ASM85 options are:

-c - [C]ase sensitive
C=<filename> - Specify [C]ode file
-F - Generate [F]ull listing
-I - Generate [I]ntel format HEX file
L=<filename> - Specify [L]isting file
P=<length> - Set page length
-Q - [Q]uiet mode
-S - Generate [S]ymbol table
-T - Output to [T]erminal
W=<width> - Set page width

For a full listing with symbol table, use -FS

Regards,
Dave

--
Dunfield Development Services http://www.dunfield.com
Low cost software development tools for embedded systems
Software/firmware development services Fax:613-256-5821

.

Relevant Pages

  • Re: Restoring a NorthStar Horizon, problems with SRAM board
    ... A-D FDC is a memory mapped IO device so I would expect some behavior ... I know my SRAM board has decoding problems. ... SRAM on the Vector Graphics Flashwriter video card located at $F000 ...
    (comp.sys.northstar)
  • Re: Fast string operations
    ... > is why people use unsafe code now and then to use pointer arithmetic to ... > loop over arrays without all the unnecessary bounds checking. ... don't return the trimmed string". ... The customer perceives this as a memory leak. ...
    (microsoft.public.dotnet.languages.csharp)
  • Re: Performance Improvement of complex data structure (hash of hashes of hashes)
    ... Anno Siegel wrote: ... >> Here is the code that I'm using to build up this data structure. ... loop through and increment the ... I know that memory allocation in C is expensive, ...
    (comp.lang.perl.misc)
  • Re: Cost of calling a standard library function
    ... It accesses/reads memory using esi 4 ... > safly move it within the cache, without having to go via ebx. ... try it the same thing on a different earlier CPU, ... should check it out...for "tight inner loop" stuff, ...
    (alt.lang.asm)
  • Re: Fast string operations
    ... worried about string performance. ... > is why people use unsafe code now and then to use pointer arithmetic to loop ... > I'm aware that looping has to occur one way or the other, but with Trim, ... The customer perceives this as a memory leak. ...
    (microsoft.public.dotnet.languages.csharp)