MicroVax II

Ooh, those are (my impression!) substantially rarer than DEQNAs. (I know _I_'d cheerfully swap a DEQNA for Qbus SCSI.) Yes. Recent (and some not-so-recent) versions are broken, in that they can't self-host; as far as I know nobody knows exactly what's wrong. My impression (as someone who hasn't tried it, but who has seen it discussed on port-vax) is that something breaks somewhere in the compiler when running native. You may also find recent(ish) versions are too resource-hungry; the MicroVAX-II can't have more than 16M RAM, which is pretty tiny by modern NetBSD's standards (pretty much ever since they relegated most ports to second-class, er, sorry, "organic" status). Fortunately, older NetBSD is still available; I think it's even available from netbsd.org. I don't have any VAXen up at the moment (I'm more constrained than historically usual about how many computers I have live or almost-live), or I'd offer to build you a tarball; I can do that anyway, but I won't be in a position to test whether it actually works, that it may be suboptimal. I imagine so, but I don't actually know, since I haven't gone looking for any. (I have fond memories of my VMS days, but as long as it remains closed-source, I'm not running it.) Same way you would any other system. In your case, I see four options: tapes, floppies (I think the RX50 is a floppy drive?), netboot (if you can find a network for the thing), and putting the disk on another machine and plopping the install on it there. Personally, I'd be installing NetBSD, and I'd probably dig out a spare DEQNA, netboot, then install onto local disk from the netbooted system. In extremis, I might download a grappling-hook program via memory binary deposit commands, then ship stuff over the console serial line. This would work, but would be slow; I think the fastest the console serial can run is 19200, or maybe 38400. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B