Dungeon Master Forum

EDIT: I've moved the download link.

The original post stating what DSB is and hopes to do follows:

With the proliferation of clones out there, every new project should probably add something new and different to the equation, so my main goal for this project was customizability, or, to put it more simply, hackability. To that end, I have embedded Lua in the engine, and just about everything that happens can be hooked and altered. Indeed, large portions of the engine itself are already written in Lua, all of which can be modified by the designer.

It is, of course, in 640x480 truecolor (why not, in this day and age) so much nicer high-res graphics can be used, but currently I'm just using the DM graphics we all know and love. That means there's not anything particularly new visually, but here are some screenshots nonetheless:

• Title Screen
  Gameplay
  More Gameplay
  It's dark!

Oh, Wow!

I don't see an emoticon for amazement.
Surprise was the best I could do.

oh wow Sophia, you told me before on a flashchat. I would use that emoticon:
could you tell something more about your project? That would be cool if it had it`s own info in encyclopedia

So that's why you needed those Anti-Fire values...

I take you're constructing the clone on real DM mechanics instead of guessed ones.

Awesome. I looked at the pics before I read your post and thought, okay, nice job, but what is the point of making an exact clone of DM if we already have the original, and then I saw Lua mentioned and all was clear. Pretty exciting stuff for dungeon makers. Speaking of which, does it support other/older dungeon formats in addition to your new content?

Paul Stevens wrote:Oh, Wow!

Were it not for your expert reverse-engineering of DM, I'd never have even attempted this.

Adamo wrote:could you tell something more about your project?

I'll confess that time spent documenting features is time that can't be spent writing code, so I've sort of been leaning more towards the latter-- a good thing, except that nobody can read my mind.

Suule wrote:I take you're constructing the clone on real DM mechanics instead of guessed ones.

As best I can, which is why CSBwin and its ability to trace just about everything interesting happening in the dungeon has been so much help. There are some formulas that just seem to be obviously bugs on FTL's part-- as the values in the code are just plain impossible or nonsensical-- and for those I am sort of doing a "best guess" as to what was intended. Of course, the game's formulas are all implemented as Lua functions, too.

Tom Hatfield wrote:does it support other/older dungeon formats in addition to your new content?

Not natively, but here's the interesting part (at least to me): Much like RTC, dungeons are compiled from a text file, but, since it's all Lua based anyway, DSB's text file is itself a Lua program. This means that though the simplest form is to just spit out a static dungeon, it could concievably spit out a randomized dungeon, or act as code to import a dungeon in another format.

good job, nice to see you working on something new

Hear ye hear ye! Sophia, you must be totally crazy!!:twisted:
But really cool' and the title is so funny !
..Dungeon Strikes Back.

Can't wait to play it

Parallax wrote:Actually, that same amount of effort will normally only give you 80% of the missing 20%, i.e. 16%, putting you at 96%. If you triple the effort you'll get 80% of the missing 4%, for a total of 99.2%. And so on and on, until you finally give up and leave good enough alone.

At least, if not more so!

By way of reference, it only took me about 3 months to get fully playable DM and CSB dungeons into RTC, but another year to iron out most of the bugs the in maps. It's now a further 5 years down the line and people are still finding minor issues...

Of course Sophias clone will be much quicker to produce as she can rip the graphics directly from RTC (or use the tools that are now available for the original games) and has the DM source code to look up all the stats / equations etc rather than having to work them all out by trial and error - but still it's certainly true to say that getting something working is only a tiny step along the way to completing it!

In fact — this is totally off-topic now — I think it's sensible to use a design pattern that takes into account this 90/90 principle.

Take Blizzard for example. They build a working engine as quickly as possible and then expand it throughout the project. That way they can test things as they go and improve the game incrementally, rather than aiming for that mystical golden milestone that so many people try and fail to reach. Considering Blizzard has produced some of the most solid games in history, I'd say this philosophy works at least part of the time.

Of course, there's no substitute for a good design document. Sitting down and writing out the details of your project beforehand will save you immeasurable grief. I'd suggest spending no less than a month on your design doc. Show it to your friends, see how they react. For reference, Warren Spector's team spent six months on the design doc for Deus Ex before Eidos told him he needed to get rolling on production. Result: one of the most acclaimed games of all time.

I agree with Tom here. Things "added on the fly" or "added in the last minute" did more wrong than good. There need to be some solid goals to achive, rahter than working chaotically.

George Gilbert wrote:she can rip the graphics directly from RTC

And how do I do that? Taking screenshots?

You refused to code a graphics exporter and ignored my offer to do it for you.

This is rare for me to say it but it's necessary: Please don't start that again.

We've been down that road once and as far as I'm concerned, and the forum in general, we resolved it. If you still have issues that you'd like to get off your chest, please do it in private emails with the people concerned.

Don't worry, I don't want to go there any more than anyone else does.

I have a new project now, it's fun, life is good.

Thank you very much for your efforts! Hoping to play your game/engine before long!

Since it's been about a month here's a brief update-- I haven't been able to do as much as I liked due to various distractions (isn't that always the way!) but I have been making good progress nonetheless--

Recently I finished implementing most of the DM monsters, freeze life boxes, and game loading and saving, so things are coming along.

Does this mean that you have already implemented the 'paint the tiles/walls' part, which to me means a major milestone.

If you mean what I think you mean (I'm not quite sure what you mean) then yes, that has been done for quite some time.

It happened to me once, in all my testing

The record/playback feature in CSBwin saved me
countless hours of debugging. When testing, I always
produced recordings and almost every 'once-only' kind
of problem was immediately reproducible via
playback.

Only just found out about this!
Very interesting work Sophia, I guessed you might do something like this.
I'll have to have a look at this lua stuff.

-- A headache makes your magic slower.

Does it now?
Hehe

Another thing...
I think you might have adept and artisan the wrong way round.
Will changing the order in xp_levelnames in global.lua cure it?

TheMormegil wrote:I think you might have adept and artisan the wrong way round.
Will changing the order in xp_levelnames in global.lua cure it?

Oops, you're right.
Yes, that'll fix it.

Since we've got a shiny new forum, I'm locking this generic topic.

Please post any new discussion in more specific threads.