Stuffing most of ToME 2 into one dungeon

[Therem Harth]

True... Now what would be nice in that case is having stairways to branch dungeons appear randomly in certain level ranges, as opposed to on specific levels like in T2.

[bio_hazard]

It's one of the things I think DCSS does really well, although there you NEED to get to the bottom of some number of branches. If you've already got a pretty good kit in ToME2, they'd be entirely skipable. And wouldn't need to replace inline special levels or themed levels in the main dungeon.

[Therem Harth]

Interesting idea. It sounds like you are thinking of a completely in-line dungeon, but have you considered having short branches for some of the special themed areas? You'd still avoid the wilderness problem, while keeping more of the variety, and allowing some choice for players to pursue the optional areas or not.

Actually that had occurred to me! Thanks for reminding me. I personally prefer the idea of randomly placed special levels, but failing that, branches would work.

(Assuming I ever get moving on this project...)

[bio_hazard]

Interesting idea. It sounds like you are thinking of a completely in-line dungeon, but have you considered having short branches for some of the special themed areas? You'd still avoid the wilderness problem, while keeping more of the variety, and allowing some choice for players to pursue the optional areas or not.

[Therem Harth]

Ah, thanks. I'll admit that using 2.3.10 as a base cuts out a lot of work (especially bugfix backporting!). When (and if) possible, you'll have to brief me on how to do a static compile against jansson.

[AnonymousHero]

However, I'm somewhat trepidatious about using 2.3.10 as a starting point. Windows is already a tangled mess when it comes to software development, and the combination of CMake, MinGW, git, and now jansson makes things even more irksome. I'm personally quite happy using Linux, but I would like to keep the Windows side of things friendly.

AFAICT it should be possible without too much effort to use a bundled statically compiled jansson. Since someone did already provide patches to get T2 compiling on Windows, that should really be the only remaining problem, AFAICT.

Unfortunately I'm swamped at the moment and don't have any way of testing Win32 builds, so I haven't tried to do it.

Anyway, I'm just saying that it will almost definitely be easier to fix the minor build problems of T2.5.10 on Win32 than it will be to work with the code base of 2.3.5. For example, stripping out all the module-dependent stuff would be completely trivial and inlining things rather than going through convoluted callback paths would be much easier, etc. etc.

(As a side note, I don't think it's ever going to be pleasant to build any cross-platform C code on Win32, regardless of what you try. To me the only real option for Win32 users who aren't also developers is to have some type of automatic CI build. I can't remember if Travis CI supports building on Win32, but it might be an option)

[getter77]

Yep, I've seen the T2 unto T4 port come on along of late---progress in all forms is welcome and the more the merrier.

Ultimately, it comes down to a case of "whatever works, works" with the finer details and points sorted as they may.

[Therem Harth]

While I'd quite like to see the entirety T2 carry forward in a more modernly accessible form for posterity---if this is what can at least get done and well situated "sooner" or more readily then definitely go for it. Besides, a full T2 conversion could always happen down the line and be better for it in terms of roadblocks cleared ahead of time by going about this mega-dungeon.

Zizzo is working on a port to the T4 engine. Pretty cool, if you have the necessary GPU power... Unfortunately a lot of computers don't (or don't have adequate drivers).

My only advice if you're actually going to do this based on ToME source would be to use my plain C port of ToME 2.x. It should be much easier to delete stuff from that.

That said, ToME 2.x has never been a pleasant codebase to work with.

If I haven't already thanked you for porting the codebase to C... Thanks! I'm glad you handled that, because I wouldn't know where to start.

However, I'm somewhat trepidatious about using 2.3.10 as a starting point. Windows is already a tangled mess when it comes to software development, and the combination of CMake, MinGW, git, and now jansson makes things even more irksome. I'm personally quite happy using Linux, but I would like to keep the Windows side of things friendly.

[AnonymousHero]

My only advice if you're actually going to do this based on ToME source would be to use my plain C port of ToME 2.x. It should be much easier to delete stuff from that.

That said, ToME 2.x has never been a pleasant codebase to work with.

[getter77]

While I'd quite like to see the entirety T2 carry forward in a more modernly accessible form for posterity---if this is what can at least get done and well situated "sooner" or more readily then definitely go for it. Besides, a full T2 conversion could always happen down the line and be better for it in terms of roadblocks cleared ahead of time by going about this mega-dungeon.

[Therem Harth]

Awww, player summoning was the best part of ToME 2. Remember GreyCat´s awesome pacifist ditl ? I wonder if that got lost when the forums collapsed.

I never saw that DitL. Anyway player summoning stops being fun after a while; you just sit around and hope your monsters don't breath on you, and eventually move in for the kill (if you're dealing with a unique).

While I dont know much about coding, given all the recent handling improvements of vanilla, wouldnt it be best to start with that and add tome like features ?

It would be best, certainly, but probably well beyond my current capabilities.

(That said, it might not be a bad idea. ToME's schools-based magic system is not my favorite, and it would be nice to really revamp the Vanilla system. Main problem is that V's magic system is kind of hardwired and rigid when it comes to adding realms.)

This actually reminds me a bit of FAangband dungeon-only mode, although not in every respect.

As far as coding difficulty goes, I would guess that depends on how deeply the wilderness is coded into T2. Actually, tracking back through the versions of ToME, Pern and Z in the variant repository might help

Yay, thank you!

I have a dream of doing something a bit similar to Unangband. Take all the rich and athmospheric content and build a small, tight, tactical dungeon crawl around it. Maybe some day!

I think Un does have a single-dungeon campaign.

(My main problem with Un is that traps are ubiquitous and even deadlier than in ToME. I've had characters instakilled by gas traps and spiked pits on level 1.)

[Mikko Lehtinen]

I have a dream of doing something a bit similar to Unangband. Take all the rich and athmospheric content and build a small, tight, tactical dungeon crawl around it. Maybe some day!

[Nick]

This actually reminds me a bit of FAangband dungeon-only mode, although not in every respect.

As far as coding difficulty goes, I would guess that depends on how deeply the wilderness is coded into T2. Actually, tracking back through the versions of ToME, Pern and Z in the variant repository might help

[Estie]

Awww, player summoning was the best part of ToME 2. Remember GreyCat´s awesome pacifist ditl ? I wonder if that got lost when the forums collapsed.

While I dont know much about coding, given all the recent handling improvements of vanilla, wouldnt it be best to start with that and add tome like features ?

[Therem Harth]

So I'm thinking of revamping my T2 fork as a single-dungeon game. I like T2's variety and (admittedly fragile) extensibility; but I've grown to dislike the huge wilderness, the multiple dungeons, and everything else that makes the game more tenuous and more time-consuming. Also I just want a single-dungeon game that's more varied than V and more forgiving than Hellband.

What I'd like to do:

- Get rid of the wilderness map. Town level is shanty town/siege camp on the surface. Ideally the townspeople would be mostly friendly and mostly fellow adventurers. Also ideally, weak (but still hostile) monsters would sometimes come up the stairs and spawn in town, to add a little danger.

- Once below the surface, all camaraderie disappears. Everyone wants Morgoth's crown, for good or ill, so everything you meet underground will be hostile. On a related note, all player summoning (and all charm spells) would be removed.

- All (non-special) levels would be small, with a relatively number of monsters.

- Monster breeding would be completely removed, to prevent the shallower levels from getting unmanageable.

- The Angband special levels from T2 would be kept, and perhaps Deathwatch and Orc Town. The others wouldn't make it in. All special levels would have a random chance of appearing within a certain depth range, and would only appear once. Princess and Fumblefingers quests would be removed.

- Floor traps would appear only in vaults, special levels, and otherwise unusual areas. There would be no random floor traps whatsoever. However, floor traps would remain invisible and lethal, to make players think a bit before taking on that vault... (And the monster "create traps" spell would still work too.)

- I'd like to implement themed levels a la Quickband. Some ideas include
* Aquifers with aquatic monsters
* Graveyard levels with appropriately tough undead
* Dragon strongholds where everything is huge, like in T2's Lonely Mountain dungeon
* Garbage dumps with all manner of slimes and jellies (and maybe zombies)
* A really nice one would be coal seem levels, complete with noxious gases, flammable dust, and a tendency to collapse when the spells start flying. That would take more serious coding though.

- There would be no monster leveling. Period. Full stop. However, some high-level monsters (most notably Balrogs) would be modified for enhanced deadliness.

- Likewise for the endgame, Sauron and Morgoth would be revamped to be more dangerous (especially Morgoth). Tiksrvwhatever and Melkor and the whole Void plotline (and any other plotline) would be removed. It might also be cool to add an uber-endboss on dlevel 127...

How does all this sound? Am I biting off more than i can chew here? Are there any single-dungeon modules still around, that I could examine for clues on where to start?

Finally, is ToME 2 even a good bet for this? e.g. how hard would it be to extend the NPPAngband magic system, with realms possibly containing much fewer than 8 books?