Date: Mon, 07 Jan 2002 12:58:28 To: John Robertson <jrr@flippers.com> From: "Fredric L. Rice" <frice@skeptictank.org> Subject: Re: 9010 compiler

At 11:03 PM 04/01/2002 -0800, you wrote:

...

Thanks for releasing that, you'd be surprised how many of us are still using or trying to figure out the use of this great tool. We use ours for checking older video and pinball game logic boards.

That seems to be the industry that's got interest. I didn't know that anybody else was using the thing.

...

One trick I am trying to solve is how to generate checksums instead of the weird (HP format) signature that Fluke elected to use as their ROM check tool. Have you any thoughts on this or, perhaps, a simple solution.

It's not so simple yet yes, there's a solution. You can locate the Z80 code inside of the F9010a that generates the CRC. If you locate the address of the entry point of the subroutine, you could then look for the jump table that exists in the machine which is used to vector to when CRCs are requested and you can install your own routine.

The CRC code is in ROM so you couldn't use my compiler to insert a Z80 checksum subroutine on top of the Fluke's CRC routine but you _could_ change the address in the Fluke's jump table to point to a _different_ routine that you could install into the box using my compiler.

Finding the jump table entry would take time since you would need to locate the CRC subroutine, note the starting address, and then locate the jump table in RAM. Then you can use my compiler to create a Z80 checksum routine, then you can change the jump table entry to call that routine instead of the real CRC.

It's tricky but it can be done.

...

I have only just started to read your documentation.... perhaps I should just read it closely and get back to you later. Thanks!

No problem! Glad to help.

...

Oh, are you familiar with the other compiler for the 9010A? It does some of what yours appears to do, from within a single DOS based program, but the documentation for it was essentially non-existent. I can email you a copy of it if you are curious.

I played with the compiler that Fluke sells and was annoyed that it was so simple. There's no real difference between what their compiler does and what's possible from the keyboard and the only real benefit of their compiler is the ability to have comments in the code. That's a _BIG_ plus since turn-around makes documenting source code so valuable.

So I wrote my own and thought I might sell it to Fluke. I got back a note telling me that since it had no sales research, no product research, and wasn't created in house, there was no interest in it.

...

Ii have a small mail-list for technical tools, that tends to be mostly about the 9010a, called techtoolslist @flippers.com, if you want to join (and be asked a few questions) it's techtoolslist-request@... subject: Subscribe. Again, thanks. John Robertson :-#)#

I'll subscribe, yes. I don't use the F9010a any more but if my compiler is going to make the rounds, that would be a good forum for answering questions and seeing if anybody has bug fixes or changes or features they might want.

At the same time I also designed and built a TTL chip tester and wrote a library for the 74LS series TTL chips that worked with my compiler. My device was basically a board with a 40 pin socket for the Z80 pod to plug in to, a 4MHZ clock source, some bi-directional latches, and a zero-insertion-force socket for the TTL chip under test.

My device routed the 16 address lines of the Z80 pod to every pin of the zero-insertion-force socket, isolated by some gates. It also routed the 8 data pins to al 16 pins of the zero- insertion-force socket. Using my compiler I could read and write addresses addresses which would drive the inputs to the TTL chip and read the outputs of the TTL chip. I wrote software that drove counters to count up then read the results to give a PASS/FAIL indication on the Fluke 9010a.

I have those 74LS programs for my compiler yet I suspect I would be hard pressed to find the schematic to my TTL tester. If I can find it I'll enter it into an ORCAD schematic and I'll make it all available.

If you want, you can dump this message into the mailing list.

To UNSUBSCRIBE from techtoolslist, send a message with "UNSUBSCRIBE" in the message body to: techtoolslist-request@flippers.com. Please direct other questions, comments, or problems to jrr@flippers.com.