9815 Test ROM 98140-66501 ?

[John Ioannidis]

The service manual for the 9815 mentions a test ROM, part number 98140-66501. A Google search for that returns a thread of this forum from 2007.

Has anyone come across that module? Are there schematics or code listings available anywhere, maybe in one of the other HP collectors' web sites that have somehow escaped indexing? It will be trivial to build one with modern parts, but I need a starting point!

[John Ioannidis]

bump

I was about to ask this question, and found my own query from two years ago

[artag]

It's also mentioned in the source code in the 9815 patented indexed on Eric Smith's site : notably in the fitst few instructions

LDS #ISTACK ; INITIALIZE STACK POINTER
LDA A, $1002 ; GET POSSIBLE TEST ROM
CMPA #$39 ; IS IT THERE ?
BNE NOTEST ; BRANCH IF NOT
JMP $1025 ; ELSE LET IT HAVE CONTROL

So a ROM decoded at $1000 having $39 (ascii 9) at offset 2 and an entry point at offset 25 will be executed with no dependency on RAM or any other i/o.

This could be something like a 98136A with nonstandard address decode and a special rom - this would then allow test/debug over the serial port.

Some other useful information would therefore be usage notes for the test rom. It wouldn't be hard to write a functional replacement. In fact an off-the-shelf 6800 rom monitor such as SWTBUG, perhaps with a few 9815 specifics such as ram and peripheral tests would probably do most of what's wanted for enthusiasts.

There's relatively little internet activity regarding the 9815 - nothing like the excellent 9825, 41 or 71 resources. But I keep coming across occasional individuals who have them, and its status as the last RPN desktop or the first industry-standard processor stands out a little. Who else has one in their collection ?

[Massimo Gnerucci]

(12-28-2022 07:10 PM)artag Wrote:  Who else has one in their collection ?

I have one, non functional, alas.

[Etienne Victoria]

Have one too.

Works except for the tape unit I have to fix (EOT detection out of order)

Need to find the time to disassemble and fix it...after I have fixed my HP 46 printer :-)

It's actually the second one I have owned (sold the first unit after restoration). I hate fixing tape units but can't resist when I find a nice machine. Printer is ubernoisy but the display and keyboard are awesome.

Best regards

Etienne

[artag]

In fact, the service manual does indicate what the test rom does - it has self-contained tests for ram, keyboard, display, tape drive etc. rather than the more general debug features I had in mind. Further tests such as printer were provided through a test tape so I guess the rom was used to get core functionality to work then the taope-loaded tests for anything else.

Many years ago I used some commissioning software built for PCs in both a rom and disk-loaded version that allowed simple repetitive operations to exercise memory and peripherals. A macro language was used to assemble more complex tests and the technicians were even able to build a hacky overlay system that copied blocks of memory to and from disc to store the saved macros.

It all seems a bit overcomplicated for a handful of machines maintained by enthusiasts : there seems to be a much larger amount of interest in the 9825. But the amount of effort involved may be quite small. I'm looking at building a simple rom board (essentially a 98134A, which is the rom board used in all the other adapters, supported by other boards for their unique i/o needs). Could then be programmed with a monitor, or a 6800 Forth, to allow me to dump other roms.