Home Page Forums MAME MAME Requirements?

The Rainbow Venix disks were feared lost forever. I've managed to read them all and am processing them into the proper image files right now. It was the only true Unix than ran on the Rainbow, and should make a nice addition to MESS'. There's indications there might be source on these, but I think that's just a re-used label. There is a c compiler, so reconstructing sources may not be outside the realm of the possible, at least for some items.

My venerable Rainbow, though, is showing signs of failure. It flakes out after it's been powered on for a few hours. Still trying to figure out why and what my next course of action should be.

So I've pulled together the ROMs and am trying to run mame on my macbook pro. I have a couple of questions. They are rather newbie-ish, so please forgive me.

I can find lots of information on floppy formats. I think I can find one that will work there. I have a bunch of floppies for Venix I'd like to try that are pure physical sector copies. What I'm less sure about is the hard disk format. I have two full disk images from a MFM emulator / reader that's read in my ST-251-1 and a friend's RD51. The emulator is at http://www.pdp8online.com/mfm/ in case you are interested. Should I just use chdman, or something else?

Second, is the serial port supported? Is that what bitbang is?

Thanks!

Warner

BTW, I've tried booting the recent venix boot disks. It's a bit slower than the real hardware, but not terrible.

However, there's some issues, but I'm not sure the root cause. It only detects 128k of RAM, but that may be due to a configuration issue on my part. It shows 128k in setup.

I create a disk with 'chdman createhd -chs 1024,6,16 -c none -o venix-rb.chd', and when venix goes to access it during install, it seems to hang. I haven't formatted it, so that may be part of the issue. The failure mode on real hardware with unformatted drives is an error about not being able to access a specific C H S.

You can find the Venix disks that I'm using at http://people.freebsd.org/~imp/venix86r-2.0.tar.xz. Root password is 'gnomes' if you get farther than I did. I've installed these on a real Rainbow 100B, so I know the images are good. The .flp files are raw images with no headers. vboot.flp and all xfer disks are bootable.

Btw, I'm running this on mac with 'mame64 rainbow -flop1 flp/vboot.img -hard1 venix-rb.chd' on a system that I'd previously gone into setup and saved.

These Venix disks recently surfaced. I've imaged them, installed them on my Rainbow, etc. I'll be blogging the experience and would like to have a section on running this under mame/mess.

Thank you for making this gem available!

I wrote a PM with some answers and a description of the current state.

The Rainbow driver should be restructured next, best before the 100-A ROMs are available.

For now, a recompile cycle is necessary (since all assumptions concerning the 100-A are kept in hand-crafted precompiler macros, which is kind of safe but very ugly).

Hint: i could use some help
Last time i tried to use MACHINE_CONFIG_FRAGMENT(.. ) for alternate machine configurations around MAME 0.160, it caused problems after the second machine reset.
Additionally i seemed unable to cover subtle differences in port logic with this approach...

It looks like I'll have to figure out how to do the recompile cycle myself. It's been, let's just say 'frustrating', the couple of times I've tried in the past.

Warner

MAME's not actually that hard to build - it's a bit easier on *IX and friends (BSD, Linux, Mac), but it's alright on Windows once you get past the initial setup.

And "make SOURCES=src/mame/drivers/rainbow.cpp REGENIE=1 -jX" will build only the Rainbow and the central core, so things go relatively quickly.

Looks like building is easy on my mac. FreeBSD fails to build. I'll build on my mac. That's something. I have somewhere to hack.

Now if I can just figure out why the 'd' key doesn't cause a 'd' to show up. It isn't limited to the Rainbow emulator. The at386 was suffering form it in the 0.184 image that I was running. About 1 in 20 times the 'd' key produced an effect

Are you sure your 'd' key is working ?

If it's a non-US/non-QWERTY keyboard that might cause problems, but in that case 'd' would never work, not 1 in 20.

All the other times I hit the 'd' key, I get 'd'. It's only in MAME running on my MAC that there's issues.... My Mac has a US qwerty keyboard, but the caps-lock is mapped to control. Maybe it's Karabiner, I'll check that out next... (exiting Karabiner fixes the problem)