Home Page Forums MAME MAME Requirements?

A word of warning: i cannot vouch for the correctness of the above DEBUG instructions. You might have to alter the M(ove) statement.

The RAM range (0100 ...) given is possibly occupied by DOS

Last time i tried in our emulation, the dump was littered with incorrect bytes (hell, i can't even rule out bugs in the emulated hard disk interface).



At least the 100-B uses an internal BIOS checksum, so we'll soon find out if the dump is correct. Hope there is time for a second shot, just in case

@bsdimp: please check your messages.

@R.Belmont & O.Galibert: i made a partial disassembly of the Z80 part (located at the end of the Rainbow-100 B BIOS).

Sector reads & writes are handled there - as far as the BIOS is concerned.

Perhaps we find out what's wrong with the timing?

https://dl.dropboxusercontent.com/u/37819653/BANNISTER/Z80_DISASSEMBLY__2016_02_04.html
(you may want to search for WRITE_SECTOR)

Thanks for having a look.

I don't suppose there's a central repo (ala github or similar) that contains the accumulated disassembly of these ROMs?

Warner

I don't know of one, but I think it'd be great if someone wants to set one up. It's not like HP will care.

@bsdimp: heard that developers are about to remove BSD support (unless somebody jumps in). For more details please have a look at the personal message sent to you.

We are still interested in the 100-A ROM - partly because older software like DOS 2.05 misbehaves. It could well be the ROM :-)

I could provide the most current ROM / Z80 disassemblies in return, together with IDA sources.

There are no plans to put them into version control, so i'd be glad about a mirror site...

https://github.com/shattered/retro-bios

PRs welcome.

For now, I made a pull request concerning the main README. It would be helpful if we had a separate directory for DEC disassemblies (*.html output from IDA).

Binary files can't be uploaded to Github, so i am a bit concerned that the IDA base file gets out of sync with future changes (to the HTML).

Do you have an advice where to put them?

How big are these binaries? Git LFS should handle them in any case -- https://github.com/blog/2069-git-large-file-storage-v1-0

@shattered: please create a folder "dec100b", because i can't...

I recently uploaded the Z80 disassembly (from February) to the repository available under https://github.com/shattered/retro-bios.

The binary for that code can be found at the end of the upper, language specific ROM.


If a Z80 guru has a look at it, i'd be happy. I still think the code tries something weird we don't emulate.

At times, the Z80 controls the 8088 (CP/M) and sometimes it's the other way round (DOS). So it's not a strict master-slave relationship.

It could also be a much simpler problem with floppy emulation (overruns when writing because of unemulated WAIT states), who knows?

Unemulated wait states causing data overruns/underruns on a machine with a 17xx FDC would be borderline fraud on the part of the hardware engineers. It makes sense on the Apple II where the software *is* the FDC, but the 17xx has a ton of ways you can sync without counting cycles.