Well, real life came calling and I have to put MAME on hold for a while. Pushed most of my WIP tree to github (https://github.com/shattered/mame/commits/gitwip), key parts are not committable unfortunately.

What's in there:

- K1801VM2 core (= severely hacked T11 core). This chip has no MMU.
- parts of chipset for this cpu (DEC DL11-compatible serial line unit, etc.)
- new bus (DEC Q-bus with add-ons to support uknc.cpp)
- new driver, ms1201.cpp (Electronika MS1201.02 single-board computer)
- de-skeletonized uknc.cpp, sans sound and floppy
- much improved DVK KCGD -- intelligent color graphics card, 800x480 px
- qbus devices:
-- DEC PC11 paper tape reader (no punch)
-- DEC RX11 floppy controller with RX01 drive (HLE)
-- virtual hard disk driver (not a real hardware device)
-- DVK KGD -- dumb monochrome framebuffer, 400x286 px
-- DVK KZD -- MFM harddisk controller (HLE, read-only)
-- UKNC ROM carts
-- UKNC IDE harddisk controller
-- skeletons of MX: and MZ: floppy controllers

What works and what doesn't:

- ms1201: 5 of 6 tests in ROM pass (remaining test is for unimplemented printer port)
- ms1201: boots and runs RT-11 SJ/FB monitors, XXDP+ small monitor
- ms1201: passes DEC VKAA, VKAB tests but fails VKAH (VKAC is supposed to fail). Passes equivalent tests 791401, 791402 but fails 791403, 791404
- uknc: passes ROM tests but fails many standalone tests
- both drivers run many games (text-only and graphical), see screeshots and videos earlier in the thread
- there's no softlist but a long TODO list for all of the above smile

There's also SIMH WIP tree at https://github.com/shattered/simh/tree/newwip -- based on Seth Morabito's 3B2 WIP, it adds simulation of AT&T DMD 5620 terminal, good enough to run GEBACA (screenshot).