HEPAX-0C Prototype version ROM images.

[Diego Diaz]

Hi all,

As announced on a previous thread, I have received what ended up being a HEPAX prototype.

After, proper cleaning and repair, it is back to life and shown its contents on CAT 2: -HEPAX 0C

I want to thank Robert Properi and Sylvain Côté for not just allowing, but encouraging me to share this finding with the 41 users community.

An archive containing the ROM images, disassembly listings and the .HEX file to emulate this prototype on a NoV-64(d) module, is available for download.

Here.

It is indeed a very early development stage version.

Hope you enjoy playing wit it. ;-)

Best from Spain.

Diego.

[Diego Diaz]

Hi all,

As requested by the owner, here you are the CAT 2 listing of the HEPAX 0C prototype:

-HEPAX 0C
HAPPCHR
HAPPREC
HARCLRC
HASROOM
HCLFL
HCRFLAS
HCRFLD
HDELCHR
HDELREC
HEPDIR
HEPROOM
HFLSIZE
HGETR
HGETREC
HGETRX
HGETX
HINSCHR
HINSREC
HPURFL
HRCLPT
HRCLPTA
HRENAME
HSAVER
HSAVERX
HSAVEX
HSEC
HSEKPT
HSEKPTA
HUNSEC
CLRAM
CODE
COPYROM
DECODE
DECODYX
DISASS
HEPAX
HEPAXA
HEXEDIT
HPROMPT
RAMTEST
RAMTOG
READROM
ROMTEST
WRTROM
XF
XFA

As you can see, there are a number of differences with the production units and some function names were changed afterwards (some of them "fortunately" changed: e.g "DISASS" -> "DISASM" :-D

Although it has less functions there are a couple of them, that were removed from production modules: RAMTEST & ROMTEST.

To no one's surprise, RAMTEST checks if the RAM page "n" addressed in "X" register is ok . It shows "n:NO RAM" if pointed to a no RAM page. On a good RAM page it shows "n:RAM TESTING" (for quite a while) and then "n:RAM OK". It shows "n: WRT PRTCTED" if the RAMTOG commad has protected that page.

ROMTEST shows the code of the ROM in page "n" addressed in "X" register (in reverse sequence if there are several banks in that page: e.g. 4, 3, 2, 1 for HEPAX) and then CHECKSUM OK or CHECKCUM ERROR depending on the case.

Best wishes.

Diego.

[Sylvain Cote]

Thank you very much Diego for your patience, for your dedication and for doing this amazing restoration job.
Thank you very much Robert for finding these modules and managing this whole project.

Diego,

Bringing back to life these dead modules show us how great your technical skills are.

Knowing that the last shipment from you took 2 months to arrive at my home, I will have to be patient before playing with this prototype marvel.
In the meantime, thanks again to you, I am now able to load the ROM image into one of my NoV-64d modules and play it with.

With the recently updated USB-41 hardware, the updated Clonix Configuration Utility and the new PIC programmer, one can only hope for new brain-blowing things from you. 8-)

To all HP-41 owner,

I you do not have a NoV-64d module, you should buy one for yourself.
The NoV-64d module can be programmed as a Clonix-D module or as a NoV-64 module, so it is a two in one package.

• The Clonix-D module is an advanced "EPROM Box" using Flash which can be configured as one unified block of 48K (12 x 4K modules) or as two blocks of 24K (6 x 4K modules) with block activation using even/odd port sensing.
• The NoV-64 module is an advanced "HEPAX" module using FRAM (64K) and Flash (48K) which can be configured in multiple ways including the obvious HEPAX mode but with a twist as it can act as four 16K RAM Advanced HEPAX modules or as two 32K RAM Advanced HEPAX + Double HEPAX Memory Modules.
  

... and to top it, it can be used in any HP-41: C, CV, CX, fullnut, halfnut and including the legendary 41CL.

Web links: Diego Díaz Clonix/NoV Web Site and my HHC 2020 presentation on his modules.

Sylvain

edit: typos

[KimH]

I dropped a note to Steen who developed the code @VM - let’s see if he has something to add.

(09-06-2020 03:17 PM)Diego Diaz Wrote:  Hi all,

As requested by the owner, here you are the CAT 2 listing of the HEPAX 0C prototype:

-HEPAX 0C
HAPPCHR
HAPPREC
HARCLRC
HASROOM
HCLFL
HCRFLAS
HCRFLD
HDELCHR
HDELREC
HEPDIR
HEPROOM
HFLSIZE
HGETR
HGETREC
HGETRX
HGETX
HINSCHR
HINSREC
HPURFL
HRCLPT
HRCLPTA
HRENAME
HSAVER
HSAVERX
HSAVEX
HSEC
HSEKPT
HSEKPTA
HUNSEC
CLRAM
CODE
COPYROM
DECODE
DECODYX
DISASS
HEPAX
HEPAXA
HEXEDIT
HPROMPT
RAMTEST
RAMTOG
READROM
ROMTEST
WRTROM
XF
XFA

As you can see, there are a number of differences with the production units and some function names were changed afterwards (some of them "fortunately" changed: e.g "DISASS" -> "DISASM" :-D

Although it has less functions there are a couple of them, that were removed from production modules: RAMTEST & ROMTEST.

To no one's surprise, RAMTEST checks if the RAM page "n" addressed in "X" register is ok . It shows "n:NO RAM" if pointed to a no RAM page. On a good RAM page it shows "n:RAM TESTING" (for quite a while) and then "n:RAM OK". It shows "n: WRT PRTCTED" if the RAMTOG commad has protected that page.

ROMTEST shows the code of the ROM in page "n" addressed in "X" register (in reverse sequence if there are several banks in that page: e.g. 4, 3, 2, 1 for HEPAX) and then CHECKSUM OK or CHECKCUM ERROR depending on the case.

Best wishes.

Diego.

[SteenPetersen]

Cool - I had forgotten all about that - well it was a few years back ;-)

I remember I had to squeeze the code to make room for all the functions we wanted to be in the shipping product - 16k is not that much - so RAMTEST and ROMTEST had to go.

I may still have a couple of disks with (very) old development images - provided they still can be read by the disk drive which has not been powered up in more than 30 years...

I think I still have the very old wirewrap boards which Sten Vesterli built the first hardware prototype on. If I find it, then I will post a picture or two ;-)

Regards,
Steen

[Diego Diaz]

Hi all,

Wow!... That was quite of a praise on your behalf, Sylvain. Sincerely appreciate your words. :-)

Steen, it would be amazing to see the "craddle" prototype from which HEPAX was born.

Needless to say that every HEPAX realted document/material you may find (and willing to share) will receive the warmest welcome from our small community.

Thanks a lot (for you and the VM team) for producing this outstanding module.

It was obviously a far more difficult task to create it with the available technology of the late eighties than it was emulating it with early 21th's. Admirable!

Best wishes and keep healthy.

Diego.

[SteenPetersen]

All,

I managed to find my box with "stuff" in the attic - and I found this:

   

This is the rest of the box - I guess there are a lot of interesting goodies in it ;-)

   

Regards,
Steen

[SteenPetersen]

Well, there is actually a lot of stuff in the box - here is a picture of some of it. I also have some good books and more, but time is running out tonight. I will try to go through it over the next days (weeks?!?) - perhaps there are something you have not seen before...

Regards,
Steen

   

[Diego Diaz]

Hi Steen,

Seems you've found the treasure chest! ;-)

I've noticed that the EPROM version in your (heck-of-a-breadboard) prototype is the "1B", seems that you kept using it after the first prototype chips from Fujisu were available.

Which is pretty reasonable at design stage since the breadboard allows full access to any point and signal while the (CPLD?) MB671209U is just a "black box".

BTW, if some spare MB671209U are still available I'd love to design a new PCB with some "improvements" (may be non-volatile RAM?) :-D

Look forward to seeing what other "jewels" you find inside the chest.

Best wishes from Canary Islands.

Diego.

[Ángel Martin]

(09-06-2020 09:04 PM)SteenPetersen Wrote:  I remember I had to squeeze the code to make room for all the functions we wanted to be in the shipping product - 16k is not that much - so RAMTEST and ROMTEST had to go.

Well, they're back now, so it may be time for a "HEPAX: The Prequel" module with unearthed jewels like those two, and perhaps adding some of the new Chain management functions as well (HEPCHN, HEPCHN? and RLSRAM)

(09-06-2020 09:04 PM)SteenPetersen Wrote:  I may still have a couple of disks with (very) old development images - provided they still can be read by the disk drive which has not been powered up in more than 30 years...

Worth the try, let's see if that HP quality lives up to its legend ;-)

Best,
ÁM

PS.- Diego, pls. remind me: is WPTOG supported in the NoVo family?

[SteenPetersen]

OK - a little update.

I managed to get hold of a set of batteries for the HP-41 - my old battery pack leaked and was thrown out at least ten years ago when I packaged the "stuff" into the box. After cleaning the battery connectors it sort of came to life with a strange looking display. Changing to normal clock speed and I got "MEMORY LOST" - so now I got a working HP-41! It has not been powered up for more than 30 years, but it still works. I enabled the double-speed again (wired by KimH many moons ago - thanks Kim) and no problem - everything appears to be normal (well - in double speed).

So next up - inserting the ADV HEPAX module - the ONLY one I got. Nothing happened - I tried all ports with normal and double speed - still nothing... This means I am basically stuck without any means of getting to my stored images...

Then I thought - does the old prototype breadboard still work? - or will it fry my HP-41??? - well I had to try. Plugging in the cable - holding my breath and turning on the HP-41 - it looked normal. CAT 2 and HEPAX 1B came up!!! So now I do have something to play with ;-D

To be continued...

Regards,
Steen

[Diego Diaz]

Hi all,

Ángel, good to see you around. :-)
I think you meant the "RAMTOG" command, used to protect RAM pages from unwanted overwriting. Yes it works in all the NoV family.

On the original NoVRAM and NoV-32 it works exactly as it does in the real HEPAX.

In the NoV-64(d) it keeps track of the protected RAM pages (active or not) and keeps its protection status as long as the module is plugged and power is present.

Sylvain found a few weeks ago that the "RAMTOG" command was not funtional in the latest version of the NoV-64(d) code (NoV-64-H.asm). I found the bug in an error I made while compliling an upgrade around 2011! (Obviously not a very demanded function) :-D

Now it is again restored and the latest version includes that correction, along with the page oriented BS. Details on the included NoV-64 manual.

Steen,
Amazing that the "mother of all breadboards" is still working after all these years! Congrats. ;-)

As a side note, if you want me to (try to) fix your HEPAX module, just let me know. I think you deserve a working HEPAX... and sure enough the whole community will second me on that.

Please consider this a compensation for the countless hours of joy that your work has brougth to many of us along decades. :-)

Best to all and keep healthy.

Diego.

[rprosperi]

Thanks for the update Steen. It's really interesting to hear you go through the rediscovery so many of us have experienced too.

Keep the updates coming...

BTW - I happen to know a guy that probably can fix that module if it doesn't come around... (and it seems he's faster than me!)

[SteenPetersen]

So next step is to check out the mass storage and the media I got.
The good old tape drive seems to be working - after charging the old battery for an hour I was able to get the directory of one of my tapes:
ASM3A
ZENROM
CCD1A
CCD2A
PPCROM1
PPCROM2
DEVEL1
DEVEL2
PLOT1A
PLOT2A
EXTIO
AUTOST
SM1C
MATH2A
XF1B
PLOTTER
BOOTROM
DISASS
So far so good - this was all the pre-HEPAX stuff.

So next up the HP 9114 tape drive - where I have two disks with these labels:
- HP-41 ROM IMAGES (likely the images seen above plus some more)
- HP-41 VM MCODE (this must be it!)

I have a modified 9114 with a W&W Software Products GmbH power supply/battery modification. I plugged in the power supply - nothing - no red LED on the power supply. Opened it up and found the fuse to be toast. Easy to fix (short circuit), but still no red LED... Found my lab power supply - set it to 6 V and connected it (I checked the schematics (found on-line) on where to connect it) - no red LED on the 9114. Measured 6V on the output of the battery pack, so the 9114 is getting the right voltage - even with 3A available from the power supply.

Any suggestions on what to do next? - are there any loose connectors or fuses inside the 9114 which could cause this?

Thanks in advance
Steen

PS: Diego, I will likely accept your offer, but I will do a little more exploring in the "treasure chest" first.

[Diego Diaz]

Hi again,

This "BOOTROM" found in your tape directory has triggered my curiosity... Do you (or someone else) know where does it come from?

Sorry about the problems with your HP-9114A disk drive.

You will more likely get better assistance if you open a new thread requesting help to fix it.

This thread is too specific and most users/readers won't stop by (or read top to bottom).

Sorry not being on much help on that, I have never dug into such devices.

On the other hand (if you happen to have a PC with a standard 3.5" floppy drive) you may use "LIFUTILS", available here.

This utility alows you (among other things) to read LIF formated discs (like those from your 9114) on a standard drive and save the content into MS-DOS type files.

Regarding my service offer, it's obviously open to whenever you want... After all these years some weeks/months up or down won't make any difference. ;-)

I do have a question regarding HEPAX internal organization of the 10 bit words inside the standard 8 bit memory chips (RAM & ROM): I know they use only 5 bits (D0-D4) of every byte, which is the most logical arrangement BTW. But can you tell me which order are those bit placed inside every memory (even/odd) bytes? Provided you remember such details :-D

Keep up enjoying your re-found treasures!

Best regards from the Canary Islands.

Diego.

[KimH]

Steen,

I happen to visit in CPH this coming weekend, I could throw a (working) 9114 w/Good battery & Charger into the bag and drop by. If you're around - I'll drop you a PM.

(09-14-2020 08:20 PM)SteenPetersen Wrote:  I have a modified 9114 with a W&W Software Products GmbH power supply/battery modification. I plugged in the power supply - nothing - no red LED on the power supply. Opened it up and found the fuse to be toast. Easy to fix (short circuit), but still no red LED... Found my lab power supply - set it to 6 V and connected it (I checked the schematics (found on-line) on where to connect it) - no red LED on the 9114. Measured 6V on the output of the battery pack, so the 9114 is getting the right voltage - even with 3A available from the power supply.

Any suggestions on what to do next? - are there any loose connectors or fuses inside the 9114 which could cause this?

Thanks in advance
Steen

PS: Diego, I will likely accept your offer, but I will do a little more exploring in the "treasure chest" first.

[Erik-M]

I recently refurbished two battery packs from two new 9114B drives.
It has a lead acid battery that is dead for sure if the drive was stored for more than a few years.
there is also a fuse in the battery pack.

I know for coincidence because of my recent repairs that W&W modified the charger so it can deliver 3 A instead of the 500mA from the original charger.
If the battery has an open circuit the drive would have worked with the W&W charger but not with the original but if it is shorted the fuse in either the pack, the power supply or both are blown. Its a solder in fuse that looks like a resistor but easy to repair, just use a normal 3A fuse with some wires soldered on.

[hth]

(09-14-2020 08:20 PM)SteenPetersen Wrote:  So next up the HP 9114 tape drive - where I have two disks with these labels:
- HP-41 ROM IMAGES (likely the images seen above plus some more)
- HP-41 VM MCODE (this must be it!)

How was the MCODE developed, is there machine readable source code?

[KimH]

Coming Sunday Steen and myself will try reading the floppy-disk with my working 9114B. Hopefully the files on the VM-MCODE disk are readable and intact.

As far as I recall Steen was developing the code on his 41 - lots of code on a small FormFactor.

Hopefully there's more news sometime on Sunday PM (European time)

(09-16-2020 04:52 PM)hth Wrote:  

(09-14-2020 08:20 PM)SteenPetersen Wrote:  So next up the HP 9114 tape drive - where I have two disks with these labels:
- HP-41 ROM IMAGES (likely the images seen above plus some more)
- HP-41 VM MCODE (this must be it!)

How was the MCODE developed, is there machine readable source code?

[SteenPetersen]

Please note I have created a new thread for the HEPAX treasure chest ;-)
- and when I get some time I will also provide some insights in how the HEPAX module came about.

I have checked with Sten Vesterli and he is not able to locate the original drawings of the HEPAX module - so we must assume this information is lost. We may be able to backtrack memory chip connections on the breadboard if this is important.

KimH came with a working 9114 drive and we determined my drive was non-functional - as his battery did not bring my 9114 back to life...

All the code was made by hand - first written on an A4 sheet of paper and when I thought it was complete, then I entered it on the HP-41 via the ASM3 assembler. I will write about this another day ;-)

Regards,
Steen