HP-41 RAMBox Questions

[rprosperi]

I recently acquired a W&W RAMbox from a European seller that was reducing his collection via TOS; after an unusually long shipping delay, it finally arrived a week ago and I started to explore it. I am trying to understand this device to explore later '41 developments but have run into many dead-ends, I'm not sure of the state of the device, and have several questions.

But first, some facts about the device and what I've found so far:

Device has a genuine W&W label "RAMBOX", but no hint if this is a 32K or 64K model.

The switches are set as follows:
A: Pages 8 & 9 are selected
B: All pages are OFF (so acting like ROM)
C: All pages are ON (enabled)

Doing a CAT-2 (in a Halfnut CX) reveals these modules (with info from PG?):

-RAMBOX 64b (Page-8, 31.45)
-KATASTER 1 (Page-9, 9.64) Presumed to be some custom module created by prior owner
-ING-VERM 1 (Page-B, 10.57) Presumed to be some custom module created by prior owner
-ING-VERM-2 (Page-D, 19.24) Presumed to be some custom module created by prior owner
ES RSU- 1A (Page-E, 4.62)

I have located and read the following resources:
a. RAMBox Owner's Manual for 32K unit (2 versions, essentially the same, from TOS)
b. Addition to 41CW and 41CY Manuals (from TOS, describes the extra "PG*" commands in 64K units)
c. "HP-41CY re-initialization ROM" thread here in the Forum, by Jurgen Keller (Thx for the hint Sylvain!)
d. hp41cy-techinfo.pdf, also from Jurgen (excellent resource!) from github site

I've read the above materials, and fully admitting some of this has still not penetrated my thick skull, I have a far better understanding of this than a week ago.

On to the questions:

1. Does anyone have a RAMBox 64 manual they can share, or point to where a copy is located? The seller has one to eventually provide, however he is quite ill and unable to do so for some time.

2. How do I verify if this RAMBox is working correctly? For example, from reading the above materials I believe if I XEQ "PG<>", the CAT-2 listing should change from "RAMBOX 64b" to RAMBOX 64a" but it does not change, and the CAT-2 listings also do not change.

3. Since the device is set to bank B, I would expect to only find modules in pages 9, B, D, and F (odd pages) yet "ES RSU- 1A" seems to be in Page-E (even).

4.a. Is "ES RSU- 1A" simply the name of the OS in the RAMBox, or is it another, totally separate module? The commands appear to be related-to, but all are different from, those seen in RAMBox 64x; also I see references to the "ES RSU 1A module" in the QRG.

4.b. There are also numerous references to "RSU modules", and some commands pertain to "all connected RSUs", implying more than one connected at a time. Does RSU (maybe RAMBox Storage Unit?) mean a RAMBox Page, or an entire device?

4.c. If a full device, did these RAMbox units come in other form factors (such as external box, connecting by port plug, like early MLDL boxes, etc.)?

5. What is the intended purpose of the A switch, since it appears the device OS is always installed in Page-8?

6. When I do CAT-2 on my RAMBox 64b, the results imply some commands appear to have been removed or disabled, with several commands replaced with "*" in the FAT; here is the listing:

*, CLLSTFL, CLPG, CLRFL, COPYPG, CRDIR, *, CRFLDTA, *, ENDPG, *, FRBYT?, *, *, GTREG, GTREGX, GETRGXY, INITPG, *, *, *, LDPGM, LDREG, LDREGX, LDREGXY, PG?, PGSUM, PTCT, READPG, SETPRV, UNPTCT, WRTPG, XQ>XR, PG<>, PG01, PG10, PG00, PG11, ABSP, ACAXY, ACLX, CLA-, PRAXY and PRL.

I believe all the commands from PG<> to the end are new in the 64k RAMBox. In the range up to the XQ>XR command, there are 33 commands in the FAT including the * commands, and this matches the number of commands in the 32K RAMbox manual.

The 9 'missing' commands are: BUFLNG?, CRFLBUF, CRFLKEY, FNC?, GTBF, GTKEY, KEYAS?, LDBUF, and LDKEY. These commands (not coincidentally) exacly fit the * locations when all commands are arranged alphabetically. These are KA and Buffer management functions, and unfortunately, the FNC? command, which may have helped exploring all this.

Is this 'normal' for RAMBox 64, or is this a custom hack?

7. Can anyone recommend additional resources to review/study?

Thanks in advance for any replies!

[Raymond Del Tondo]

Hi,

"ES RSU- 1A" is an Eramco Systems ROM, usually as part of their own RAM boxes;-)

AFAIK the 64k version of the W&W RAM box was named "W&W RAM-Box II",
and is covered in the CY manual.

The "RAMBOX 64b" ROM will work for 32k and 64k RAM boxes.
Of course in 32k units, the PG.. functions are meaningless...

Actually if the unit doesn't react on a PG<> ,
then I'd say it's a 32k version.

If it were a 64k version (aka RAM-Box II),
and the counterpart of "b" waited there,
it would switch pages, and show "RAMBOX 64a" in the CAT.

If the "RAMBOX 64a" isn't there, then I'd expect kinda
reaction from the machine, be it either a noticable delay
or even lock up.


Back then I revived several CY units and RAM boxes using
my CY-INIT ROM.

It worked for 32k units, too.
However up to ca. 2007, 32k units revived using my tool
have the "b" ROM on page 8, since my tool
was written with the CY as main target,
and didn't check whether bank switching was successful or not.

So first the "a" image was written, then a bank switch was issued,
then the "b" image was written.
Later I made a separate init ROM for 32k units.


However if there are holes in the CAT listing, then it may also be a modified version,
or worse, a defective image.

[rprosperi]

Thanks Raymond, this explains a lot!

As for "ES RSU- 1A", yeah, I guess the "ES" was there all along as a hint, and I now see it's also in the QRG (it's always easy to see these things after one knows the answer...).

Since my device is presumably a RAMBOX 32, which image should be restored: -RAMBOX 1B, or another, intact version of RAMBOX 64b? The extra commands in the latter presumably might be handy (excluding bank mgmt. Fns)

As I understand this, Jurgen's new tool will not work, as that is tailored to the dual 32K banks in a 41CY (or presumably also a RAMBOX II).

Is your init ROM for 32K units available?

Possibly naive new question: If your init ROM is not available, can my image be repaired by un-protecting Page-8 of my module, and using HEPAX or ZenROM plus an image of the appropriate new RAMBox image in a CLONIX module page, copying that image to Page-8, thus overwriting the present modified/corrupted image?

Thanks again Raymond!

Though Raymond's comments resolve most of my questions, I'd still appreciate replies on questions 5 & 7 if anyone else has info. to share.

[Sylvain Cote]

(02-17-2020 01:57 AM)rprosperi Wrote:  Possibly naive new question: If your init ROM is not available, can my image be repaired by un-protecting Page-8 of my module, and using HEPAX or ZenROM plus an image of the appropriate new RAMBox image in a CLONIX module page, copying that image to Page-8, thus overwriting the present modified/corrupted image?

Yep that's it, since you do not need HEXPAX to do this, you can configure your NoV-64d module as a Clonix-D and copy the ROM's in it.

(02-17-2020 01:57 AM)rprosperi Wrote:  Though Raymond's comments resolve most of my questions, I'd still appreciate replies on questions 5 & 7 if anyone else has info. to share.

Q5: You can use the RAMBox32 as a standard MLDL box and when used that way you do not need the RAMBox OS at all.

Q7: Nop! what you have listed is what I have.

Most of what I know comes from these documents, trial and error testing several setup, hanging my units several times until I understood them completely.
Jürgen initialization module was invaluable for restoring my hanged units while testing.

Sylvain

[rprosperi]

(02-17-2020 03:09 AM)Sylvain Cote Wrote:  

(02-17-2020 01:57 AM)rprosperi Wrote:  Possibly naive new question: If your init ROM is not available, can my image be repaired by un-protecting Page-8 of my module, and using HEPAX or ZenROM plus an image of the appropriate new RAMBox image in a CLONIX module page, copying that image to Page-8, thus overwriting the present modified/corrupted image?

Yep that's it, since you do not need HEXPAX to do this, you can configure your NoV-64d module as a Clonix-D and copy the ROM's in it.

(02-17-2020 01:57 AM)rprosperi Wrote:  Though Raymond's comments resolve most of my questions, I'd still appreciate replies on questions 5 & 7 if anyone else has info. to share.

Q5: You can use the RAMBox32 as a standard MLDL box and when used that way you do not need the RAMBox OS at all.

Q7: Nop! what you have listed is what I have.

Most of what I know comes from these documents, trial and error testing several setup, hanging my units several times until I understood them completely.
Jürgen initialization module was invaluable for restoring my hanged units while testing.

Sylvain

Great, that's all good news, thanks for confirming these points.

I mentioned ZenROM and HEPAX as they both have utility functions for copying ROM. I believe I'll have to use some different ROM for this, even though the RAMBox contains the COPYPG function, as the RAMBox ROM image will be replaced, so that function can't be run from here (right?).

I'll email you the exact proposed procedure just to double-check, as I don't want to kill my RAMBox completely and would like a nod to confirm it; as it is now, it's still mostly functional. I doubt the details of the process interest anyone else here.

[Monte Dalrymple]

(02-16-2020 08:17 PM)rprosperi Wrote:  Doing a CAT-2 (in a Halfnut CX) reveals these modules (with info from PG?):

-RAMBOX 64b (Page-8, 31.45)
-KATASTER 1 (Page-9, 9.64) Presumed to be some custom module created by prior owner
-ING-VERM 1 (Page-B, 10.57) Presumed to be some custom module created by prior owner
-ING-VERM-2 (Page-D, 19.24) Presumed to be some custom module created by prior owner
ES RSU- 1A (Page-E, 4.62)

If the custom module images are at all interesting I can add them to the CL image.

[rprosperi]

(02-17-2020 06:03 AM)Monte Dalrymple Wrote:  If the custom module images are at all interesting I can add them to the CL image.

I will do my best to save images of the 3 custom modules if the RAMBOX image isn't too corrupted. They appear to be collections of FOCAL programs for engineering (?) assembled by a prior owner, rather than vintage 3rd party modules, but I can't tell for certain as the programs I inspected are in German. As such, I will contact the guy I purchased this from to verify sharing is OK. Unless he created these himself and objects, I'm happy to send them for inclusion in the CL.

[HP-Collection]

I do not know if Raymonds INIT-Rom is available, but you have also the possibility to use Jürgen Kellers Init ROM....

RB2ini

Matthias

[rprosperi]

(02-17-2020 06:38 PM)HP-Collection Wrote:  I do not know if Raymonds INIT-Rom is available, but you have also the possibility to use Jürgen Kellers Init ROM....

RB2ini

Matthias

Yes, I'm familiar with that thanks, however Jürgen's solution is for the 41CY (or a RAMBOX II w/64KB RAM) and the technique used (though really, really clever) does not apply to 32KB RAMBOX I units, which it appears I have.

The Forum article and docs in the github project however provided LOTS of useful information and background and inspired me to give this a go, rebuilding my RAMBOX OS page using a CLONIX module with ZenROM and a copy of intact RAMBOX 64b image to re-write Page-8.

[MeindertKuipers]

I do have a copy of the manual, an English version and a Gereman version, I have put them in my DropBox here: https://www.dropbox.com/sh/i21k3giahao6z...Im1Ma?dl=0

[rprosperi]

(02-17-2020 07:09 PM)MeindertKuipers Wrote:  I do have a copy of the manual, an English version and a Gereman version, I have put them in my DropBox here: https://www.dropbox.com/sh/i21k3giahao6z...Im1Ma?dl=0

Thanks Meindert. The English one is the same as on TOS, but this German one appears to be new to the community sites, you may want to send a copy to Warren to add to his collection.

Thanks for sharing!

[J-F Garnier]

(02-17-2020 08:26 PM)rprosperi Wrote:  

(02-17-2020 07:09 PM)MeindertKuipers Wrote:  I do have a copy of the manual, an English version and a Gereman version, I have put them in my DropBox here: https://www.dropbox.com/sh/i21k3giahao6z...Im1Ma?dl=0

Thanks Meindert. The English one is the same as on TOS, but this German one appears to be new to the community sites, you may want to send a copy to Warren to add to his collection.

That English manual is actually a (apparently personal) translation of the German manual. The official English "RAMBOX Owner's Manual 1.1" is on TOS too (I provided it some years ago with the CY supplement). Look for all W&W Software entries.
And the German manual is on TOS too.

J-f

[rprosperi]

(02-17-2020 09:08 PM)J-F Garnier Wrote:  That English manual is actually a (apparently personal) translation of the German manual. The official English "RAMBOX Owner's Manual 1.1" is on TOS too (I provided it some years ago with the CY supplement). Look for all W&W Software entries.
And the German manual is on TOS too.

J-f

Yes, both English versions are there. Comparing them a few days ago, with hints from Jeremy's letter, I concluded the English version from W&W was not immediately available (at least not in the US) so the guy Jeremy mentioned undertook translating it and then shared it. It is mostly accurate, but there are several typos, including in some function names, which initially made me think the functions may have changed, but I eventually concluded it was simply typos.

I thought the German manual on TOS was only for the 41CY, but I see it is for both the 41CY and RAMBOX II.

RAMBOX Manual Summary:

The German manual on TOS is for 41CY and 64K RAMBOX II
The German manual from Meindert above is for 32K RAMBOX I
The early English manual on TOS (and above) is for 32K RAMBOX I
The later English manual on TOS is for the 32K RAMBOX I
The English addition on TOS is for 41CW and 41CY (but portions also pertain to 64K RAMBOX II)

The only manual of this series that is missing is an English manual for 64K RAMBOX II, although most of that content can be gleaned from the other manuals available.

However, I am still puzzled about the extra commands for 64K RAMBOX.

The German manuals mentioned above explain the PG<>, PG01 and PG10 commands, but the ROM in my device includes these 11 commands (beyond the same 33 from the RAMBOX I set):

PG<>, PG01, PG10, PG00, PG11, ABSP, ACAXY, ACLX, CLA-, PRAXY and PRL.

Perhaps the prior owner customized this ROM adding these additional utility routines?

[Sylvain Cote]

(02-18-2020 12:29 AM)rprosperi Wrote:  The German manuals mentioned above explain the PG<>, PG01 and PG10 commands, but the ROM in my device includes these 11 commands (beyond the same 33 from the RAMBOX I set):

PG<>, PG01, PG10, PG00, PG11, ABSP, ACAXY, ACLX, CLA-, PRAXY and PRL.

Perhaps the prior owner customized this ROM adding these additional utility routines?

I have listed the functions for W&W ROM images related to HP-41CW, RAMBOX, HP-41CY & RAMBOX II in Jürgen "HP-41CY re-initialization ROM" thread.

These functions are related to HP-41CY / RAMBOX II who has 64K of QROM (2 x 32K, named RAMBOX #0 and RAMBOX #1)

Included in the manual

• PG01 map RAMBOX #0 even pages (8, 10, 12, 14) and RAMBOX #1 odd pages (9, 11, 13, 15)
• PG10 map RAMBOX #1 even pages (8, 10, 12, 14) and RAMBOX #0 odd pages (9, 11, 13, 15)
• PG<> switches between PG01 and PG10 settings
  

Missing from the manual (educated guess)

• PG00 map RAMBOX #0 even pages (8, 10, 12, 14) and RAMBOX #0 odd pages (9, 11, 13, 15)
• PG11 map RAMBOX #1 even pages (8, 10, 12, 14) and RAMBOX #1 odd pages (9, 11, 13, 15)
  

These functions are coming from the CCD ROM

• ABSP : alpha backspace (OM P6.29)
• ACAXY : accumulate alpha and X by X (OM P6.17)
• ACLX : accumulate line by X (OM P6.22)
• CLA- : erase alpha from the right (OM P6.30)
• PRAXY print alpha and X by Y (OM P6.20)
• PRL : print line (OM P6.24)
  

[rprosperi]

Thanks for your normal thorough detective work Sylvain. I figured out the easy ones, playing around, but it did not occur to me to check for other W&W products as the source.

Reading old RAMBOX threads last night, I did see other references to PG00 and PG11, so perhaps these came in later releases of the RAMBOX OS, or even user-written, it seems they are relatively simple (for an MCODER) sequences of ENROM commands.

Examining some of the FOCAL program code in the custom ROMs reveals they definitely do use some (all?) of these 'extra' commands, so it appears a prior owner copied these utility functions from the CCD ROM to this hacked ROMBox OS image.

[Raymond Del Tondo]

(02-18-2020 03:53 AM)Sylvain Cote Wrote:  Included in the manual

• PG01 map RAMBOX #0 even pages (8, 10, 12, 14) and RAMBOX #1 odd pages (9, 11, 13, 15)
• PG10 map RAMBOX #1 even pages (8, 10, 12, 14) and RAMBOX #0 odd pages (9, 11, 13, 15)
• PG<> switches between PG01 and PG10 settings
  

A small correction to your description of PG<> :
According to the manual PG<> switches between RAMBOX 0 and RAMBOX 1,
so if page 8 contains RBOS-"a", all pages are switched to the "b" after issuing PG<> ,
regardless whether banks were interleaved before or not.

(02-18-2020 03:53 AM)Sylvain Cote Wrote:  Missing from the manual (educated guess)

• PG00 map RAMBOX #0 even pages (8, 10, 12, 14) and RAMBOX #0 odd pages (9, 11, 13, 15)
• PG11 map RAMBOX #1 even pages (8, 10, 12, 14) and RAMBOX #1 odd pages (9, 11, 13, 15)
  

Not needed, since PG<> completely switches to RAM Box 0 or 1.

This is what I noted back then when analyzing the CY OS:

Code:


*        ENBANK functionality in the RAMBOX:
*        --------------------------------------
*        ENBANK1: reset all pages to bank A
*        ENBANK2: enable odd pages of bank B
*        ENBANK3: enable even pages of bank B
*****************************************************************************
    .NAME    "PG<>"

***Bank A
[8DBC]    ENROM2
        ENROM3

***Bank B
*[8DBC]    C=0    ALL
*        ENROM1

        RTN

*****************************************************************************
    .NAME    "PG01"
[8DC3]    ENROM1                    ; Reset all pages to bank A
        ENROM2                    ; enable odd pages of bank B
        RTN

*****************************************************************************
    .NAME    "PG10"
[8DCA]    ENROM1                    ; Reset all pages to bank A
        ENROM3                    ; enable even pages of bank B
        RTN

[Sylvain Cote]

Hello Raymond,

(02-18-2020 09:12 PM)Raymond Del Tondo Wrote:  A small correction to your description of PG<> :
According to the manual PG<> switches between RAMBOX 0 and RAMBOX 1, so if page 8 contains RBOS-"a", all pages are switched to the "b" after issuing PG<> , regardless whether banks were interleaved before or not.

You are right and I was mistaken.

I just did the following test on my HP-41CY that proves what you have said.

Activate RAMBOX a/0 non-interleave

Code:

PG<>
CAT 2                                    // if shows "-RAMBOX 64b" then do another: PG<>

Create odd and even modules in RAMBOX a/0

Code:

CAT 2                                    // shows "-RAMBOX 64a"
11  ENTER  "-TEST-AO"  INITPG  CAT 2     // shows "-RAMBOX 64a" ... "-TEST-AO"
12  ENTER  "-TEST-AE"  INITPG  CAT 2     // shows "-RAMBOX 64a" ... "-TEST-AO" ... "-TEST-AE"

Create odd and even modules in RAMBOX b/1

Code:

PG<>
CAT 2                                    // shows "-RAMBOX 64b"
13  ENTER  "-TEST-BO"  INITPG  CAT 2     // shows "-RAMBOX 64b" ... "-TEST-BO"
14  ENTER  "-TEST-BE"  INITPG  CAT 2     // shows "-RAMBOX 64b" ... "-TEST-BO" ... "-TEST-BE"

Switch RAMBOX a and b between interleaved and non-interleaved modes:

Code:

PG<>
CAT 2                                    // shows "-RAMBOX 64a" ... "-TEST-AO" ... "-TEST-AE"
PG01
CAT 2                                    // shows "-RAMBOX 64a" ... "-TEST-BO" ... "-TEST-AE"  // reordered for easy reading
PG10
CAT 2                                    // shows "-RAMBOX 64b" ... "-TEST-AO" ... "-TEST-BE"

Code:

PG<>
CAT 2                                    // shows "-RAMBOX 64a" ... "-TEST-AO" ... "-TEST-AE"
PG<>
CAT 2                                    // shows "-RAMBOX 64b" ... "-TEST-BO" ... "-TEST-BE"

Cleanup

Code:

PG<>
CAT 2                                    // shows "-RAMBOX 64a" ... "-TEST-AO" ... "-TEST-AE"
12  CLPG  CAT 2                          // shows "-RAMBOX 64a" ... "-TEST-AO"
11  CLPG  CAT 2                          // shows "-RAMBOX 64a"

Code:

PG<>
CAT 2                                    // shows "-RAMBOX 64b" ... "-TEST-BO" ... "-TEST-BE"
14  CLPG  CAT 2                          // shows "-RAMBOX 64b" ... "-TEST-BO"
13  CLPG  CAT 2                          // shows "-RAMBOX 64b"

Sylvain

[Sylvain Cote]

FYI,
You can have a RAMBOX II (64k) replica by using a NoV-64d (or a NoV-64) module configured with the HP-41CY simulation and when inserted into a HP-41CX Halfnut you get a HP-41CY (non-turbo) replica.
I have tested it and I get the same behavior as the original, RAM shadowing is supported but unfortunately TURBO mode is not.
Sylvain

[Massimo Gnerucci]

(02-19-2020 04:01 AM)Sylvain Cote Wrote:  FYI,
You can have a RAMBOX II (64k) replica by using a NoV-64d (or a NoV-64) module configured with the HP-41CY simulation and when inserted into a HP-41CX Halfnut you get a HP-41CY (non-turbo) replica.
I have tested it and I get the same behavior as the original, RAM shadowing is supported but unfortunately TURBO mode is not.
Sylvain

Does it also work into a DM41X? Therein speed should not be an issue.

[Sylvain Cote]

(02-19-2020 07:25 AM)Massimo Gnerucci Wrote:  Does it also work into a DM41X? Therein speed should not be an issue.

Since the DM41X does not have an I/O block to receive the module, it does not.