what happened to Meta Kernel 3.0 ?

[dmh]

I've just started using Meta Kernel 2.3 and note that the documentation mentions Meta Kernel 3.0 and gives some screenshots.

Did MK 3.0 ever get released?

If so, where do I find it?

If not, what happened?

Also, MK 2.3 includes the condensed 62 page documentation but I've seen reference to a 150 page version of the documentation. Where can I find this version?

TIA, dmh

[Massimo Gnerucci]

(03-21-2023 10:04 AM)dmh Wrote:  Did MK 3.0 ever get released?

If so, where do I find it?

You can find it in the 49G.

[dmh]

What about for the 48GX?

(03-21-2023 10:16 AM)Massimo Gnerucci Wrote:  

(03-21-2023 10:04 AM)dmh Wrote:  Did MK 3.0 ever get released?

If so, where do I find it?

You can find it in the 49G.

[Hiwi]

Hi,
MetaKernel 3.0 for the 48 was annonced, but never released.
For me 2.3 works good, are there any issues ?
Ralf

[Massimo Gnerucci]

(03-21-2023 10:21 AM)dmh Wrote:  What about for the 48GX?

(03-21-2023 10:16 AM)Massimo Gnerucci Wrote:  You can find it in the 49G.

AFAIK the developers were hired by HP (ACO), and MK v3 was included in the 49G: you can see its logo flashing by at boot time.

[dmh]

Early days but no problems. It's nice and fast :-)

MK 3.0 mentions even more speed improvements which would be great (along with other new features). Too greedy? ;-)

(03-21-2023 11:09 AM)Hiwi Wrote:  Hi,
MetaKernel 3.0 for the 48 was annonced, but never released.
For me 2.3 works good, are there any issues ?
Ralf

[Mark Power]

Quote: MK 2.3 includes the condensed 62 page documentation but I've seen reference to a 150 page version of the documentation. Where can I find this version?

Eric Rechlin has been on the hunt for the missing full manual:

https://literature.hpcalc.org/items/1266

I found this one which we think is the one you are looking for, though it is labelled MK2.10:

https://www.hpcc.org/programs/hp48/MK210mod.pdf

The page numbers only go up to 91, but you'll see from the landscape format of the document that there are two manual pages per PDF page, so really it is 181 pages long.

[dmh]

So I've found a minor issue, disassembly using ->S2 or setting MK flag -21 for SysRPL display does not use the entry marked as "Prefer".

I'm using Table/table.48g from the mk230 download for the entries and table.txt has the correct "Prefer" configuration as per below:

Code:

=ONE            EQU #03FF9 *    Prefer
=real           EQU #03FF9 *
=MEMERR         EQU #03FF9 *

However, disassembly shows MEMERR instead of ONE.

I'm guessing this may be a bug unless the binary entry file wasn't built using the text file.

Are entry files a standard format so I can try another one (from, say, Jazz)?

Also, what is Examples/ENTRIES.A for? It is similar to the above but has about 600 less lines and the binary version is not included.

Any suggestions to resolve this appreciated.

(03-21-2023 11:09 AM)Hiwi Wrote:  Hi,
MetaKernel 3.0 for the 48 was annonced, but never released.
For me 2.3 works good, are there any issues ?
Ralf

[Hiwi]

Hi dmh,
interesting. I am also using table.48g and have the same issue.

If flag -21 is set and you edit the file, do you have the same issue?
Do you have the !RPL directive in your file?

There is another table for MK at hpcalc org: extablehp48mk.zip
I will try itto check if it has the same issue.

I dont think, that JAZZ tables work with Metakernel, in the distribution you have a PC programm to compile your own table.
Ralf

[dmh]

I thought I only saw Pascal source code for the table creation?

The txt file also mentions using gentab2 which is in other downloads.

Unless ‘prefer’ isn’t getting encoded properly it may be a bug in the reading and handling of the entry table.

The Sys RPL compile works fine and the code runs, it’s just the disassembly shows the wrong command (not the one tagged with prefer).

I might try another couple of commands with a preferred label and see what happens and report back.

Let me know how you go.

(03-22-2023 11:40 AM)Hiwi Wrote:  Hi dmh,
interesting. I am also using table.48g and have the same issue.

If flag -21 is set and you edit the file, do you have the same issue?
Do you have the !RPL directive in your file?

There is another table for MK at hpcalc org: extablehp48mk.zip
I will try itto check if it has the same issue.

I dont think, that JAZZ tables work with Metakernel, in the distribution you have a PC programm to compile your own table.
Ralf

[Hiwi]

Hi dmh,
to you have a suggestion for a working memory dissassembler?
I tried RPL48, but this crashes my HP48G.
Not to diassemble code like ->ASM, rather to examine the rom entry points like =LOOP eg.
and so on.
Ralf

[dmh]

I've just tried DOB5 for MK 2.x and it works with the few basic tests I've done.

Returns SysRPL or machine code based on the command you enter. So DUPDUP returns SysRPL and DUP returns machine code.

Find at DOB5

(03-22-2023 02:56 PM)Hiwi Wrote:  Hi dmh,
to you have a suggestion for a working memory dissassembler?
I tried RPL48, but this crashes my HP48G.
Not to diassemble code like ->ASM, rather to examine the rom entry points like =LOOP eg.
and so on.
Ralf

[Hiwi]

Hi dmh,
fantastic i made some tests and it is exactly what i wanted and works without any trouble.
Ralf

[Hiwi]

DOB5 only works on Firmware R Revisions on my daily use HP48, which has revision P it does not work :-( There is to source code avaiable, so no chance to fix it.
Ralf

[dmh]

Oh no :-(

What happens / what does it do?

I only have an R ROM device, lucky I guess.

(03-27-2023 08:42 AM)Hiwi Wrote:  DOB5 only works on Firmware R Revisions on my daily use HP48, which has revision P it does not work :-( There is to source code avaiable, so no chance to fix it.
Ralf

[Hiwi]

Hi dmh,
it works in almost every case. EC, DOB, and within the editor. Highlighting and EC & name.

When editing and pressing CST and then pressing ROM: My HP aks me friendly if i like to recover the memory :-)
Ralf