drdisv4 is the missing piece. It’s a 2K utility ROM at $E800–$EFFF that adds printer support and disk utilities on top of MONA and loader41. It must sit at $E800: the loader’s commands $F7 and $F9 jump to $EE4E and $EE19 in this ROM. My guess is that the name means Drucker (printer) + Disk, version 4. DRDISV4.ASM is complete disassembled source for TASM32. When I assemble it, the result matches drdisv4.bin byte for byte, all 2048 bytes. I’ve attached the converted drdisv4.bin as well. Printer and terminal Parallel printer (Centronics-style), at $E801 (prints a CR) and $E803 (prints the character in A): It drives the 6532 at $F600, sending data on PB. PA bit 1 is the strobe. It waits for the printer on PA bits 3 and 4. Serial output, software-generated, at $E8A4: bit-bangs on PA, with the baud rate set at $E91F–$E954 using the 6532 timers. The matching serial input is at $E95A. Terminal hooks with printer echo, which wrap MONA’s own routines: $E9CD wraps TTYOUT. $E9F0 wraps CARRET. $EA08 wraps ASCIIN. Ctrl-P ($10) switches printer echo on and off, with the flag kept at $F757. Disk utilities, built on loader41 Two monitor commands, $EB00 and $EB02: $EB00 (CLC) formats a disk. It writes all 40 tracks with the WD write-track command ($F4), using the format bytes in FMTTAB at $ED90. It then reads every sector back and writes an empty directory. $EB02 (SEC) appears to copy a disk from drive 0 to drive 1, file by file. Both return to MONA’s STP and show $0000 in the address display on success, or $FFFF on error, just like MONA’s own tape ER. Loader extensions: $F9 ($EE19) sets a file’s position for random access. $F7 ($EE4E) appears to copy a file from one drive to the other. Boot loaders: $EF20 (drive 0) and $EF23 (drive 1) load the file BASIC into memory starting at $C000. $EFAA loads the file DOS to $E000. Save memory: $EF70 writes MONA’s STRTL/H…ENDEL/H range to a file using WRBYTE. About the source Equates: it uses the same names as the other files, so everything fits together: MONA’s routines and variables (CARRET, TTYOUT, ASCIIN, STP, TAB1, STRTL…). MONA4’s port addresses (PA–PBDD at $F600). loader41’s entry points and RAM names. Calls into loader routines without a name use the same L_xxxx labels as LOADER41.ASM. Labels: apart from FORMAT, FMTTAB, BASNAM and DOSNAM, labels are L_xxxx, named after their address. The descriptions above match those addresses. Hidden entry points: two are made with a skip byte ($EB00/$EB02 and $EF20/$EF23). The source writes them as .BYTE $24 and .BYTE $2C. Interrupt routine: the format code points the IRQ vector at $ED10, a short abort routine (PLA PLA PLA SEI RTS). Odd details: The bit-bang serial code uses $F687, $F695 and $F696, which are timer registers in the old $F680 6532 block rather than MONA4’s $F600. Maybe the board has a second 6532 there. $EF16–$EF1F and $EFC9–$EFFF are never used and look like leftover bytes. I kept them as .BYTE so the image matches. With MONA4, loader41 and drdisv4 together you have the complete set: monitor, disk operating system, and printer/disk utilities. Yes, drdisv4 fits MONA2, and better than it fits MONA4. The MONA routines it calls are in the same place in every version. It calls TAB1 ($F80B), STP ($F868), ASCIIN ($F9F1), CARRET ($FA4F) and TTYOUT ($FA86). I checked all three binaries (1.10.77, mona2 and mona4), and these routines start at the same addresses in each. The calls are fine with any MONA version. The timer addresses point to MONA2’s hardware. The serial routines use the 6532 timer registers at $F687, $F695 and $F696. Those belong to the $F680 6532 that MONA2 uses: $F687 is MONA2’s VORBEI and $F696 is its T_64. $F695 is the ÷8 timer, which MONA doesn’t use itself. MONA4 moved that 6532 to $F600, with its timers at $F607–$F617. Under MONA4, drdisv4’s timing would only work if there were still a chip at $F680. The $F600 port is a second I/O chip on a MONA2 board. The parallel printer and bit-bang serial use PA/PB at $F600–$F603. On a MONA2 system nothing in MONA uses those addresses, so that is presumably a second 6532 or similar port chip just for the printer. On MONA4 the same addresses are MONA’s own port, so drdisv4 would clash with it. The way results are shown fits MONA2. The format and copy commands return through STP with $0000 or $FFFF in the address, meant to be read on the 7-segment display. MONA2 still has the display and key cycle. MONA4 bypasses it and goes straight to the teletype. loader41 works with both. It only uses its own controller at $BFC0–$BFC4, its RAM at $B400–$B7FF and MONA’s zero-page pointers. So the likely set is MONA2 + loader41 + drdisv4, with the floppy controller plus a second I/O chip at $F600 for the printer. MONA4 looks like a later teletype-only system with its main port moved. In DRDISV4.ASM those three timer addresses are still plain numbers. I can rename them VORBEI, T_8 and T_64 to match MONA2.ASM; T_8 would be a new name, since MONA itself never uses that timer. The source would still assemble to the same bytes. DRDISV4.ASM now uses the same names as MONA2 for its I/O. It still assembles to exactly the same 2048 bytes as drdisv4.bin. Timers: these are the $F680 6532 that MONA2 uses. VORBEI ($F687) and T_64 ($F696) have the same names and values as in MONA2.ASM. T_8 ($F695) is a new name for the ÷8 timer, which MONA itself never uses. Printer port: the port at $F600–$F603 is now named PA2, PADD2, PB2 and PBDD2. In MONA2, PA…PBDD mean the $F680 6532, so the old names would have clashed with MONA2’s. All the other names are as before: MONA routines and variables (CARRET, TTYOUT, ASCIIN, STP, TAB1, STRTL…) and loader41’s entry points and RAM. So MONA2.ASM, LOADER41.ASM and DRDISV4.ASM now describe one consistent system