Great game... for people who like rogue likes 😝.
I tested with my commodore pc10-III, it has Nec v20 and can run in 4,77/~8/~9Mhz. For graphics I've installed for testing genoa superega card.
Game is definitely playable... bit sluggish few times while exploring message about npc battles got lost because keys for movement were already pressed when the notification came.
What I didn't like is not the game but the genre.. for example when I died I had over 4000 gold but nowhere to use the gold.
Would be nice if there were taverns to use the money, rest and re-equip, and travel between floors, doing missions for example destroying portals for tavern owners or temples, but that's not of course scope of this game.
Btw is there version of nethack that would run on nec v20?

Thank you for the feedback Kekkula! You'll be happy to hear that one of the next updates will involve what I call static levels, set in stone levels with things like towns with new NPCs, non-destructable buildings, etc. So hopefully you'll have some more changes to spend all that money! Missions / Quests is an interesting idea too, not one that I had thought of but I'll definitely play with the idea in my head.

I've had the same thing happen to me where i mistakenly cancel out of a message box because I've been playing very quickly. I was considering adding a short message queue system so you could hit something like M to see the last 10 messages.

You're welcome.
I don't know the limits of your engine or your skills so I just throw ideas... I'm not a programmer.
Message queue system or some specific key like spacebar to be pushed to make the message dissappear would be nice.

For the portals how about hiding them to hidden rooms where monsters would break out. Now that you have adlib music the ambient music could have humming in it if there are active portals still on level 🤷‍♂️

I've been playing with trying to get Adlib music playing during the gameplay, I haven't cracked that one yet because you still have to use commands to push the music on a note by note basis and there is no multithreading on these old systems. I have a few broken experimental builds that try to create a rudimentary time sharing scheme to play music and poll for input but so far they've been pretty unplayable. I've been meaning to scour ebay for more vintage game development books to find the lost arts of squeaking every last bit of performance out of the original PC hardware.

If anyone has any books they think are great fits for the era please let me know!

I hadn't considered the idea of exclusively putting generators into hidden rooms, I'll have to play around with that concept. Little hidden pockets of monsters could be a fun idea. I think you'd be very unlikely to find those rooms before all the monsters had generated and the generator had popped so that might take a somewhat big re-think of the generators. All of the new systems are things I spent a lot of time playing with but I think a lot of them need some tweaking to make sure they're a fun balance of risk / reward and still give the player maximum agency to decide how they want to proceed through the dungeon.

Great progress in v. 1.6

I like the robustness of the NPC system and the weapon system. The game has been very stable. Saves work perfectly. The equipment and weapon / armor selection is quite robust.

Your idea of hidden pockets / conclaves of monsters would add great challenge to the game. Defeating these hidden pockets of monsters could carry vast rewards of course (exp, gold, items). I greatly miss a way to spend all of that gold acquired. I think a town level with stores and even special items, the power of which, which are re-stocked based upon character level when the town is visited. Also a weapons shop, perhaps. This town level would flesh out the game greatly.

What is a reasonable max. level depth? Floor 255? I'd also welcome more and more challenging monsters as one "presses deeper..." up to and beyond floor 200, etc. I think there are about 10-12 monster types now. I think about 30 monsters is "perfect". Demon Lords, Three-headed hydras, Elementals, Devils (Flame, etc), Zombies, Demon knights, possibly some "unique" monsters perhaps even from other games with "impossible" stats like Diablo, Xyzzy, Zork, etc.

Quick technical question:

With v. 1.54 the binary was self-contained, a single executable with all the monster art and game text included within the executable file. With v. 1.6.0 the game text and monster art is now in a separate approx. 12kb text file.

Were you running into memory problems? Is it possible to compile a static binary with the monster art and text included all in the .exe, it would be much appreciated and make the game once again completely self-contained.

Is there a technical reason why you moved the game prose narratives and monster art outside the .exe?

Hi Zyzzle,

Thanks for all the great feedback!

For max floor depth, I think 255 is probably a reasonable target. The furthest I’ve personally made it so far is floor 17. There are currently 24 monsters in the game, and I plan to continue adding more as development goes on. Unique monsters and bosses are planned for 1.7.

I’m hoping to publish a 1.7 roadmap soon to give everyone a clearer idea of where the next major update is headed. One thing I want to do differently for 1.7 is avoid a single massive drop with a bunch of new systems all at once. Instead, I’m planning a series of incremental releases, roughly 1.6.1 through 1.7, with each update introducing one new system at a time. The goal there is to get new versions into people’s hands faster and to gather feedback earlier so I can incorporate community ideas more quickly.

As a bit of a sneak preview, some of the things I’m planning or actively exploring include player ghosts, smithing, towns, unique monsters and bosses, a legacy system where certain achievements unlock new functionality in future playthroughs, additional NPCs, monsters, items, and spells. I’m also experimenting with the idea of timer-based interrupts to enable music during gameplay and battles, though that’s still very much a technical challenge. Another stretch goal is a high score sharing system using QR codes to submit runs to a leaderboard.

Regarding the use of external VGM and DAT files, that’s exactly right. I started running into real-mode memory limits and had to decouple some data from the main executable. I’ll likely push further in this direction, with monster, item, and spell data moving into external DAT files as well. It feels like a more robust way to manage increasingly complex data that doesn’t need to live in RAM full-time, and it’s also very much in line with how commercial games of the era handled things.

If external files are causing a specific issue for you, let me know. There may be a middle ground I can address in future updates.

Oh nearly forgot I'm also going to try to rewrite certain slow routines like character fills on large/full screen rewrites, and dungeon generation in ASM to speed things up. I finally have my XT back together and 1.6 runs pretty slowly on the stock 8088 at 4.77 Mhz.

delinthe wrote on 2026-02-09, 18:15:

Oh nearly forgot I'm also going to try to rewrite certain slow routines like character fills on large/full screen rewrites, and dungeon generation in ASM to speed things up. I finally have my XT back together and 1.6 runs pretty slowly on the stock 8088 at 4.77 Mhz.

Thanks for the update and the progression line. I like all of those ideas. I've noticed the game takes a long time to generate the dungeon levels on vintage hardware. ASM code welcome there!

If a "modular" (piecemeal file) approach works best to develop and add new features, by all means stick to it. However, I was merely pointing out that on very old systems loading everything into RAM at one go and contained in one .exe will be much faster. I've been running the game from a floppy on an 8088, and the loading of the seperate art files and VGMs does take longer in 1.6.0 as opposed to 1.54. Less disk access for floppy system is better of course -- but this is admittedly a fringe case.

As it stands now, the uncompressed binary is about 200k and the data files add another ~25kb. You've still got tons of memory to work with on a 640k system for a single .exe. (I remember one of the last versions of Moria, 4.8.x for real-mode DOS in 1988 was released as an entirely self-contained real-mode .EXE which was 500kb! Then they had to switch to modular files as well due to memory limits of the single .exe compile.

Quick status update I am working towards version 1.7! Currently working on a build of version 1.6.3 which I'll likely post up this weekend. Added in a scoring system, high score ghosts, and boss encounters! Also for my sanity I'm working on a debug menu so I can test new content much more rapidly.

Wonderful! Getting to be a very polished text roguelike. Thanks for the update. I look forward to trying out your new version.

Hey everyone, back with another quick update!

Since the last update I've expanded the feature set and scope of version 1.7 (uh oh… feature creep!). Nine of the eleven planned phases are now complete.

The headline addition is a full town system. When transitioning to floor 10 the game loads a hand-authored static level from a binary .LEV file containing buildings, shopkeepers, and wandering NPCs. There are planned static towns for floors 20, 30, and 40 as well. For now the deeper periodic towns fall back to a procedural generator that writes a valid .LEV file at runtime. In addition to the periodic towns there are also rare random towns that can be discovered in the dungeon, which transport you to a dynamically generated town with a selection of NPCs. I've also added a Town Portal scroll that allows you to teleport to a random town at any time.

Also new in this version is a Legacy System that persists across deaths in a LEGACY.DAT file. This system currently tracks ten cross-run milestones that unlock things like the Legacy Ring, Phantom Blade, Iron Chests, and rare NPC variants.

There's also a new boss monster system with a dedicated full-width ASCII battle pane, special abilities, and per-type defeat tracking. To support these encounters I added several related gameplay systems:

• Doors (wood and iron) with stuck/locked mechanics
• A kick command with a chance of hurting your foot
• An anvil-based smithing system where cascading STR/DEF/AGI skill checks can improve, diminish, or eventually destroy the item you're working on

I've also added five new level 7 monsters to fill the difficulty gap before the endgame tier: Death Knight, Vampire, Manticore, Storm Giant, and Nightmare.

One new system used heavily in the static levels is the trigger system. Tiles can trigger events such as dialog boxes, sound effects, music start/stop, battles, flag setting, NPC or monster party spawns, monster or NPC movement, and door creation, destruction, or state changes. This opens up a lot of possibilities for scripted encounters and environmental storytelling.

After experimenting with bosses as rare random spawns, and with door-kick triggered encounters, I felt the difficulty spike from randomly encountering bosses might not be very fun or fair. Instead I've started creating hidden sub-dungeons that players can discover and enter in order to hunt down and defeat the four boss monsters. These areas use static levels and will leverage the trigger system extensively to create more thematic encounters that tie into the developing storyline.

Version 1.7 also introduces a new scoring system that makes the high score chase more interesting. The system now takes into account player level, dungeon level, inventory and equipment value, gold, and kills. This gives a much more complete picture of how a run went compared to the old system, which only tracked dungeon level with tie-breakers for kills and player level.

I've also added three new NPC types:

• A book seller who only sells magical tomes
• An enchanter who never fails to enchant items but charges more depending on the item's existing enchantment level
• A shopkeeper used in towns who allows unlimited buying and selling

The shopkeeper NPC can also have a static inventory, allowing me to create themed merchants like potion sellers, weapon stores, or consumables shops.

I've also tweaked the high score ghosts to make them more challenging since they were total pushovers in the original build. They now have three abilities (the most of any monster):

• A chance to phase through your attacks
• A drain attack that restores some of their health
• A wail attack that can silence the player, preventing singing or magic for several turns

Next up is the quest system, which I'm still designing. The basic idea is that players will be able to pick up quests and complete them for rewards. More to come once that system is fleshed out further.

Finally I'll be evaluating a few technical improvements for the engine. First is adding color support for users with EGA/VGA graphics cards, followed by controller support for players who want to use gamepads or joysticks. I'm also looking into compressing the art data to reduce disk usage since it is currently the fastest-growing asset in the build. Lastly I plan to do a full top-to-bottom code review to identify routines that could be rewritten in assembly so the game remains playable on PC/XT class hardware.