CMT71-64E HP-71b eprom module questions

[KimH]

(04-28-2021 12:02 PM)dmh Wrote:  

(04-28-2021 11:47 AM)KimH Wrote:  I don't think there were many built - doubt anyone knows. AFAIK there are about 6-7 folks (all active in this forum) who has one which works.

I am almost certain the BAD comes from the size OR that the ROM checksum generated originally by PTGENC$, is wrong or never generated before the burn.

If you found the document, look at page 15

What you can try is the command - DISP PTSIZE$ (':PORT(n)') - 32k would give h10000 or double that with 64K

I couldn't find the document you mentioned but I ran the command above and got h20000 so it appears it is a 64k EPROM with the 32k test program.

Is there a way of manually loading the 64k test?

You asked for it

71B Compendium

about halfway down you can see the "burner" and a link to the owners guide.

This 71B Compendium by Sylvain is an outstanding (!!) and complete document which will be inspiration for you for some time, if you wanna know more about your recent acquisition and where to spend your hard earned money to have even more fun.

You will also find some of the tools - like the PIL-Box - suggested by Dave F earlier in then stream. Tools you would need to get "stuff" like the 64K test Lex into your 71B - there's plenty to keep you busy on rainy days.

[Dave Frederickson]

(04-28-2021 12:02 PM)dmh Wrote:  Is there a way of manually loading the 64k test?

The file needs to reside on the EPROM it's testing, but that's a waste of time. The "Port" command PTTEST$ can verify the checksum of any EPROM.

From the HP-71B Compendium, here's the EPROM manual.
https://dl.dropboxusercontent.com/u/5323...xKE_OM.pdf

And here's the programmer manual.
https://dl.dropboxusercontent.com/u/5323...P01_OM.pdf

Dave

[dmh]

(04-28-2021 02:56 PM)Dave Frederickson Wrote:  

(04-28-2021 12:02 PM)dmh Wrote:  Is there a way of manually loading the 64k test?

The file needs to reside on the EPROM it's testing, but that's a waste of time. The "Port" command PTTEST$ can verify the checksum of any EPROM.

From the HP-71B Compendium, here's the EPROM manual.
https://dl.dropboxusercontent.com/u/5323...xKE_OM.pdf

And here's the programmer manual.
https://dl.dropboxusercontent.com/u/5323...P01_OM.pdf

Dave

Found this in the manual and tried it but it doesn't give 00000.

Programs I've tested seem to work though. Maybe the 32k test in a 64k EPROM is related (unless PTTEST$ is module independent)?

I will keep tinkering - thanks all for your helpful information and tips :-)

[Dave Frederickson]

(04-28-2021 03:05 PM)dmh Wrote:  Found this in the manual and tried it but it doesn't give 00000.

Programs I've tested seem to work though. Maybe the 32k test in a 64k EPROM is related (unless PTTEST$ is module independent)?

I will keep tinkering - thanks all for your helpful information and tips :-)

As KimH mentioned, PTTEST$ and ROMTST will only pass if a checksum is stored in the EPROM, so if the test fails it doesn't necessarily mean the EPROM is bad.

[rprosperi]

(04-28-2021 03:14 PM)Dave Frederickson Wrote:  

(04-28-2021 03:05 PM)dmh Wrote:  Found this in the manual and tried it but it doesn't give 00000.

Programs I've tested seem to work though. Maybe the 32k test in a 64k EPROM is related (unless PTTEST$ is module independent)?

I will keep tinkering - thanks all for your helpful information and tips :-)

As KimH mentioned, PTTEST$ and ROMTST will only pass if a checksum is stored in the EPROM, so if the test fails it doesn't necessarily mean the EPROM is bad.

Further, many (most?) 3rd-party EPROMs did not bother to use checksums, probably stopping once the process of creating them was understood and stable. These are not needed or used by the 71 itself, so only useful to verify manually later on.

Most likely the contents are intact, I've only seen one bad EPROM among probably several hundred (not all mine, but as reported here).

It's pretty clear the original surveying app was wiped and a prior owner simply copied a bunch of 'handy' files onto the EPROM, so they were ready to copy to RAM after a MEM LOST. Some of these files (LEX, BASIC) could be used directly from EPROM, but RPN (a BIN file) cannot, as it stores the XYZT stack values inside itself (presuming that it's in RAM) and does not work at all when run in EPROM.

What are the dates of the various files (the CAT listing is not complete) that would likely give a good hint on when this image was created? Also, what is the size of the COORD and SUBPROGS programs? Back in the day these modules were expensive (actually, still true today, but less so in relative dollars) so it's unlikely the original publishers would have used a 64K module, if the actual application fit on a 32K module. Finally, at least some of these files were commercial and not (legally) allowed to be distributed on a commercial product).

[dmh]

(04-28-2021 04:42 PM)rprosperi Wrote:  

(04-28-2021 03:14 PM)Dave Frederickson Wrote:  As KimH mentioned, PTTEST$ and ROMTST will only pass if a checksum is stored in the EPROM, so if the test fails it doesn't necessarily mean the EPROM is bad.

Further, many (most?) 3rd-party EPROMs did not bother to use checksums, probably stopping once the process of creating them was understood and stable. These are not needed or used by the 71 itself, so only useful to verify manually later on.

Most likely the contents are intact, I've only seen one bad EPROM among probably several hundred (not all mine, but as reported here).

It's pretty clear the original surveying app was wiped and a prior owner simply copied a bunch of 'handy' files onto the EPROM, so they were ready to copy to RAM after a MEM LOST. Some of these files (LEX, BASIC) could be used directly from EPROM, but RPN (a BIN file) cannot, as it stores the XYZT stack values inside itself (presuming that it's in RAM) and does not work at all when run in EPROM.

What are the dates of the various files (the CAT listing is not complete) that would likely give a good hint on when this image was created? Also, what is the size of the COORD and SUBPROGS programs? Back in the day these modules were expensive (actually, still true today, but less so in relative dollars) so it's unlikely the original publishers would have used a 64K module, if the actual application fit on a 32K module. Finally, at least some of these files were commercial and not (legally) allowed to be distributed on a commercial product).

Dates are all 01/01/00.

COORD size is 8902

SUBPROGS size is 954