Home Page Forums MAME MAME Requirements?

IIRC the only ROM completely missing which is necessary for accurate emulation from a Rainbow 100B is the z80 arbitrator PROM 23-090B1, which is an MMI6308 256x8 Open Collector PROM at location E13 (E11 on the Rainbow 100A, both share the same part). The other missing roms are bios/firmware language roms, and we have one of the 5 known sets dumped. It would be nice to dump the other four, though.


In other words:

We have the two video PROMS, 23-023B2 and 23-056B2.

We have the video PAL 23-117J5 at E88/E93 (two pals on the board, both have the same contents).

We have the 23-037E3 Character set rom.

We have the 23-022E5 (main Firmware) rom.

We have the following language roms:
23-020E5 (German, French, English)


We're MISSING the following language roms:
23-015E5 (Dutch, French, English)
23-016E5 (Finnish, Swedish, English)
23-017E5 (Danish, Norwegian, English)
23-018E5 (Spanish, Italian, English)

We're MISSING the z80 arbitrator PROM:
23-090B1


If any of the roms/proms/pals in your machine have numbers different to the ones listed above, let us know!
Note that for pals and proms, DEC usually only printed the last 5 digits of the part number on the PROM/PAL itself, so the part may have stamped on it somewhere something like "WB8146 // 058B1" (where // is a newline) for 23-058B1

We're also missing ALL of the Rainbow 100A roms: firmware, language, character set, proms, pals. everything.

LN

Nice to see another competent person on board.

First, we would be glad to have additional source code or commented BIOS listings for this system.

Even with unoffical patches applied and a lot of tweaking, the rainbow.c driver misbehaves under DOS 3 and CP/M ("write error").

I hit a brick wall when it comes to 100% accurate floppy emulation. Some keywords: consistent floppy writes, contended memory, insertion of Z80 WAIT states during floppy I/O.

Drop me a line of you are interested in an unofficial build or the header files currently needed to get a working system...

Read support for floppies, hard disk emulation, and ClikClok are already built in. But there is still a lot to do ;-)

I have the 100B. Years ago, I bought a 100A used, but along the way I parted with it. I don't recall now when or why or how, but I think it was at least 15 years ago. So I can't help there.

I found the "Shared Memory Arbitrator" on the sheets, and located the part. I was thinking that I could pop it out and put it in a reader, but no such luck. I can see why you don't have this part. It is soldered into place and I'm loathe to take it out of the circuit. The circuit around it is also a bit complicated, which may frustrate efforts to read it in sutu. Mine is tagged with "6308-1J P8346 \\ LM8438 090B1" if that matters.

This is the first time in years I've looked at the inside of my unit. It's clean and looks good. So the next step is to power it up when I can get a few minutes and round up the right cables... That will let me read the disks I have. The RX-50 will connect to a PC floppy controller, and I have a couple of machines that still have that in them if I can't get this to power up for some reason... I'm taking it slow turning it on after so long since the last time...

The stickers over all the EEPROMs on the boards are blank. I didn't write down the part numbers, though. I should have taken pics while I had it apart to examine later. If I take it apart again, I'll do that.

I have a friend that may have a Rainbow 100A. Is there a way to dump the EEPROMs that is useful from MS-DOS's debug? I'd normally expect that I could just dump a memory range. If so, I can lookup the debug commands and try them on my 100B and go over to my friend's house and snag them from him. Any suggestions / accumulated wisdom before I dive in?

Depending on the memory map you may be able to dump the whole firmware and language roms from dos DEBUG, but it is probably best to do with an eprom programmer.

Also, we'd need pictures of the chip labels anyway to determine what firmware revision the machine has.

LN

I'm pretty sure the program ROM for it is only visible to the Z80, not the main CPU. You might be able to somehow trojan it out, but that would require finding/exploiting a bug in it.

You should be able to dump the BIOS within DEBUG.COM (from DOS).

Ranges on a Rainbow-100 A are:

ROM (FA000-FBFFF) (E89) 8 K
ROM (FC000-FDFFF) (E90) 8 K
ROM (FE000-FFFFF) (E91) 8 K

The character ROM [E98] 2732 (4 K) can't be accessed this way.

Unverified:

Derived from instructions for IBM-PC computers -
http://www.mess.org/dumping/dump_bios_using_debug

It's really super easy to open the machine and take a photo, then grab the ROMs via debug. My friend may allow me top open it up and take a picture. I doubt he'd allow me to pull chips. He's doing me a favor as it is, since he's not touched this machine in 20 years except to move it...

Warner

The Z80 on the Rainbow comes up in reset. The 8088 has to put its program in it, and take it out of reset. The boot roms are mapped only on the 8088 side. In fact the only thing the Z80 is attached to, apart from the shared memory and the 8088 interrupt line is the floppy drive.

But I also have Z80 compilers and assemblers if that's helpful. I didn't see anything on the prints, though, that could easily be read back from the Z80, with the possible exception of the Z80 memory arbitrator PROM, though that's a long-shot at best given the machinations one would have to go through to toggle the different address lines and chasing down where they all go... I'm not sure they all go to something that's readable by one or the other of the CPUs.

Woot! My friend has agreed to lend me his 100A. More to follow.