What's wrong with DSB?

[Sophia]

DSB is really for programmers, not visually minded people who enjoy the RTC way of doing things. I'm really struggling to get anywhere with DSB to be honest.

This kind of disappoints me to hear. While I'll admit DSB is kind of geared to a certain mentality that I have and others may not, I have tried to make it easier to use.

I wonder what more can be done to make DSB more accessible.

[Gambit37]

I've not got anywhere yet because every time I sit down with it and try to get started, I get a mental block.

I don't think it's a question of accessibility. I think it's more that I don't want to let go of 10 years (!) tinkering with RTC. I think somehow I'm subconsciously scuppering my ability to learn something new, because I don't want to have to

Lack of time doesn't help either. I know with RTC, if I have an hour spare, I can get a few things built. With DSB, it's a complete unknown, so I guess I'm resistant to trying it for fear that I won't achieve anything in the same amount of time. It's an unknown unknown

I need to sit down, probably for a weekend, and just work through the basics. If someone could give me a completely free weekend, that would be great

Having said that...

I wonder what more can be done to make DSB more accessible

Point & Click for creating *everything*. That would be nice (I know it's impossible, don't worry, just teasing!)

[linflas]

I can code LUA, i really like this language but DSB needs an editor as intuitive as RTC Editor and also a basic TXT format like XML or similar into which you can call external LUA functions and for the engine, err... 5 squares ahead view or more, auto-resizeable bitmaps, bigger resolution, maybe very light 3D code for walls ?

[Joramun]

DSB/ESB already does a lot of things...
I guess for non-programmers, ESB lacks two fundamental things:
First: 2 features: new archetype creation, and custom content implementation.
Second: it's not self-sustained:
- it requires a lua-highlighted-editor (for comfort)
- it requires external programs to obfuscate dungeons (which some might want, but not I)

On the other hand, from my point of view: nothing is wrong with DSB.
The DM community is small, and people have invested already in a few clones (CSBWin, RTC and DMJava)
DSB, however polished, needs some time to rise as what it is: the most powerful clone.
Once a few custom dungeons are completed, and DM and CSB have been thoroughly tested,
it may boast the title of up and running clone.

I've several WIP dungeon/ideas:
- Chaos Hack (the random dungeon generator)
- An extension of the small prison
- A fantasy content pack (basically an extension of MFI's work)
- A Sci-fi content pack (similar but with Doom/Strife etc. content)
- Tons of ideas
But no time whatsoever.

[zoom]

For easier learning curve videos would be nice. --> screen recordings /
So how to set up your workspace for dsb,
what files there are and how they connect (maybe with graphic chart) Get directions(this is maybe where Gambit struggles?)
--
One video could be about the differences to original DM..
like weight influence on throwing distance (which annoyed me personally , but which, iirc can also be easily fixed)

WHy dsb in the first place - had this in mind -->resolution increase would be the thing, but we had this discussion already..

MAybe a teaser video:
Dsb has vast possibilities, like minigames e.g. play frogger to pass over to the next side of a pit or something..
like an axe that flies randomly around and changes directions/ could also work for spells etc..
and see through wall -videos/ animations in game.. (--> just record the test dungeon )
This is admittedly only for the very advanced user and for many due to programming without assistance /teamwork not doable.
But there is the forum ,where you can ask nicely aso..

EDITOR
Use DSB as your dungeon creation tool of choice. Say and show why!
Make a video of the use of the editor as it is now. THen people can comment maybe more how to improve the editor. And see how it works. It is sometimes all about getting used to
(I only used RTC editor couple of times years ago/ages ago!! Cannot comment..csbuild is the only editor I know slightly. In this regard: something like DSAs for DSB!! )

[ian_scho]

DSB, however polished, needs some time to rise as what it is: the most powerful clone.
....
But no time whatsoever.

Ditto!

If I've not said it already, I love DSB. Thank you!

[Sophia]

Thanks, guys. I actually wasn't just fishing for compliments. Honest!

a basic TXT format like XML or similar into which you can call external LUA functions

I don't understand this. There is the "function_caller" that will call an arbitrary function when sent an activate message from a button or floorpad. Is something like that what you meant?

- it requires a lua-highlighted-editor (for comfort)
- it requires external programs to obfuscate dungeons (which some might want, but not I)

This is a good point. The first thing, I think, may be able to be resolved. Scintilla, which is the engine behind the popular SciTE editor, is apparently not that hard to embed, and has a pretty good feature set. I may be able to include a Lua editor in ESB, though I wonder whether the advantages are worth the time it'd take to do so.

As for the second, unfortunately, I'm not sure what I can do. The Lua team has advised that people embedding Lua do not include compilers in their software, and they don't include an easy way to output compiled Lua. So recommending that you use the external Lua compiler is the only thing I can really do for obfuscating/compressing your Lua, other than obfuscating the source code somehow, which seems kind of silly.

like an axe that flies randomly around and changes directions

The test_dungeon actually already has one of these.

[linflas]

a basic TXT format like XML or similar into which you can call external LUA functions

I don't understand this. There is the "function_caller" that will call an arbitrary function when sent an activate message from a button or floorpad. Is something like that what you meant?

Now that I have my brain back, i think my idea is a total non-sense... I was thinking about replacing dungeon.lua with some plain text format (XML or RTC-ish) with tags for clasic components (teleporter, monster, object, decoration..) and some <lua> tag for event customization. But i think it's already doable by identating properly your lua code.

[zoom]

as a sidenote:

like an axe that flies randomly around and changes directions

The test_dungeon actually already has one of these.

I know! :O THat´s where I got it from in the first place --> I wrote a line later as a conclusion:(--> just record the test dungeon )

.. makes me wonder about the readability of my posts as a sidenote to the sidenote: Notice, I forgot(simply did not know) to add the _

[Sophia]

makes me wonder about the readability of my posts

No, I think your posts are readable. I just didn't comprehend what you were getting at.

[Gambit37]

Something else I wanted to add too:

I know DSB is more powerful than RTC and offers a lot more flexibility. However, I would be more inclined to properly try and learn DSB if it offered something more than RTC in terms of modern graphics. At the end of the day, it's still the DM engine, and that does feel quite limiting by today's standards.

In particular, I would love to be able to have a bigger gaming viewport (448x272 is pretty darn small after all), and probably with a different perspective too so that we could get more detail in the envionments and higher roofs. (Although my Zagor dungeon is old school and works fine at the old size, I've been tinkering with some higher-res stuff for another custom game. It looks a bit poo and would look so much better in a bigger viewport.)

I think it would be awesome if we could have it running in say 1024x768, (with a viewport of 800x600 (or whatever) and use the remaining space to create unique interfaces by positioning the other stuff where we want. Like in Dungeon Grind:

http://mattiasgustavsson.com/Blog/Entri ... legame.php
http://dungeon-master.com/forum/viewtop ... 19&t=28469

Maybe if you seriously considered this, you could port it as an app for iPad or something? Make some money out of it!?

[Sophia]

In particular, I would love to be able to have a bigger gaming viewport (448x272 is pretty darn small after all), and probably with a different perspective too so that we could get more detail in the envionments and higher roofs. (Although my Zagor dungeon is old school and works fine at the old size, I've been tinkering with some higher-res stuff for another custom game. It looks a bit poo and would look so much better in a bigger viewport.)

I think it would be awesome if we could have it running in say 1024x768, (with a viewport of 800x600 (or whatever) and use the remaining space to create unique interfaces by positioning the other stuff where we want.

Well, the first and foremost issue in doing something like this is simply the amount of work for me in making all of this stuff changeable and flexible, and the amount of work for someone else in making the entire new interface-- and they somewhat feed off of one another, because I can understand someone not wanting to do the work of making the new interface until there was an engine to support it, and of course I have hesitation to do the work without something that's actually going to use it. So we'd have to coordinate our efforts, which always makes projects harder.

There's a speed issue, too. Running at 1024x768 is likely to be a bit sluggish. I've mentioned this before, but I think people are somewhat in disbelief because modern computers are so fast. However, they aren't really that fast as what DSB would need to do. The thing is, DSB isn't built like a modern game at all. Like RTC, DM-PC, and a bunch of other old games, the CPU draws everything directly into a main RAM buffer, and then it is simply copied over into video memory every single frame. The CPU isn't as optimized for graphics drawing as the GPU, and the speed of the video bus (copying 1024x768 pixels every single frame!) is also a somewhat limiting factor. Modern games have all of their graphics data stored on the video card, and the only thing going over the video bus are OpenGL (or whatever) instructions telling the GPU what to do with that data, which is why they can get good framerates at crazy resolutions. To take full advantage of modern graphics I'd have to convert DSB to use OpenGL or something... which is its own big messy project...

Maybe if you seriously considered this, you could port it as an app for iPad or something? Make some money out of it!?

DSB is loaded with calls to the Windows API. A port would be tough. Even if that were not the case there is enough stuff in DSB copyrighted by FTL that I'd have quite a bit of work ahead of me regardless.

[Gambit37]

I know it's a tall order. See it as a challenge

[money]

I havn't designed my own dungeon yet - time is by far the biggest challange - but I do enjoy, when ime permits, having a go at some of the others people have created. I guess where DSB is concerned there are hardly any custom dungeons out there that get people engaged. As you know i completed the DSB version of DM and enjoyed it, a few differences (percieved differences in any case), but all in all an enjoyable and nicely created port.

If there are customer dungeons close to completion or available to play I would be more than happy to play them on DSB as much as any other port.

[Gambit37]

@Sophia: How hard would it be to extend DSB to more than a 3-tile-ahead view? I know we've briefly talked about it before and you said it was tricky -- just wondering if your views on that have changed?

I'd love to get more believable day-time outdoors views, and you really do need more visible tiles for that.

[zoom]

If we cannot increase the resolution of the viewport easily maybe we can extend it?
In this regard I was thinking about a solo-run DSB version(could also work with dm2 version from sphenx), with one single champion.
You would get rid of some Interface parts you do not need anymore(whole upper portion as well as movement keys.could use that space to extend the viewport on.
With clickable areas on monsters you could even make it touch /gesture compliant..
anyways, would getting rid of parts of the interfaces for the other 3 champions be of any use?
Sure, there must be a bullet proof concept first with much considered... also having the work involved in mind!(but as I see it , dsb can quickly move interfaces around?)
Moving the single champion position (from side to side is only needed)
I would do with 2 areas on the bottom screen.(left half of viewport width(standing left aligned) and right part of viewport width( for standing right aligned))

before I go on here I want to add :
maybe it is possible to only increase resolution for the items ? Maze of Zagor did not look that bad with mixed resolutions. Besides I think personally that the walls are ok with crude resolution as is and that the items look indeed better if they are in higher resolution.
well, I do not know the facts(what resolution dsb is currently ) but these are my thoughts right now.
I would not go too far , though. Monsters in higher than eob resolution with still conventional animation phases would arguably not be convincing.

edit:
what if Rasmus would code opengl capability? then we would have higher resolution, no?

edit2:
in all this regard we could think about how to make dsb scifi compatible..what interface elements or changes would be needed if any? etc..

[Sophia]

It was/is tricky, but not completely out of the question. It's a fair bit of work, and I'm not going to do all of it.
Things haven't really changed significantly since I wrote this bit:

In the case of DSB, it would be easier for all involved for me to just work with an artsy person to figure out some sensible default coordinates and a field of view, and implement dynamically scaled flooritems, pickableuppables, etc., out to that new larger field of view. Statically scaled flooritems as well as wallsets won't ever work beyond the initial DM range because the graphics just don't exist, and I don't feel like even going there. Fortunately, the good news is that in large outdoor wallsets, most graphics are generated via the previous two means, not the latter two.

If we cannot increase the resolution of the viewport easily maybe we can extend it?
...
maybe it is possible to only increase resolution for the items ?

This stuff is not really any easier from a program standpoint. It does allow some re-use of graphics, though. For example, an 896x544 viewport could be used with existing graphics, though many of the graphics would start to look pretty blocky.

[Mon Ful Ir]

If DSB was an add-on for RTC, everyone on this site would be using it by now. But RTC develops habits of thought, and dungeon designers are used to working with RTC's interface and around RTC's constraints. To someone accustomed to RTC, DSB seems dauntingly complex and you have to go back through that very first learning curve again -- it's like switching from QWERTY to DVORAK. You know DVORAK's better but there's so much to relearn...

I've only really played with DSB to a very limited extent, but I already know I'll never design another RTC dungeon. From my code library thing, I love the way the cockatrices and basilisks and medusae petrify you, and I love the mind flayers, and I love the underwater feature. I'll come back to that at some point but I'm distracted at the moment (there are all sorts of things going on in my life, some of which are good things).

For me there are only three dungeon-building programs that still hold my interest: DSB, the Morrowind TES construction set, and GemRB (which is certainly going to be my go-to program if they ever finish it).

[linflas]

@Sophia: How hard would it be to extend DSB to more than a 3-tile-ahead view? I know we've briefly talked about it before ....

Reading that post again makes me remember showing Sophia a demo of lua code from another engine (Demoniak3D) where a tree was seen up to 5-squares ahead. I tried to port this to DSB but I got lost into all DSB functions ...

[Soaponarope]

If DSB was an add-on for RTC, everyone on this site would be using it by now. But RTC develops habits of thought, and dungeon designers are used to working with RTC's interface and around RTC's constraints. To someone accustomed to RTC, DSB seems dauntingly complex and you have to go back through that very first learning curve again -- it's like switching from QWERTY to DVORAK. You know DVORAK's better but there's so much to relearn...

I just checked out DSB and this pretty much sums up what I was thinking. After spending a lot of time working with RTC and knowing it so well, it will be hard to make the swap for someone like me with no programming experience. Working on creative aspects is something that is very fun for me; learning Lua is not. I don't know how much I will need to learn exactly as I've only just started playing around with the basics, but from other comments it seems programming experience is needed. I am quite interested in seeing what DSB offers, though on a first look RTC's better visual editing more appeals to me.

It also looks and feels less polished than RTC while in game. Even on my powerful computer it seems to have some lag issues. The cursor slows when items are picked up and the projectile speeds are much too slow.

I'm not putting it down or saying RTC is better, just stating what I felt were the downsides from an initial look, given the nature of this thread. Some of this is a matter of personal preference, though could be how others feel when first loading it up. From playing the test dungeon it looks like some cool things can be done(and I'm sure many more) and I am glad to see an editor that pushes beyond RTC in it's limits. I'm just not sure yet if it is right for me, but will be looking into it more.

[Sophia]

Fair points... mostly. I do take slight issue with the statement that DSB projectile speeds are too slow, because they're far closer to the projectile speeds as found in authentic DM. The issue is really that RTC projectile speeds are too fast, and people have somewhat gotten used to this.

[Soaponarope]

They may be the same as the original DM, I haven't played it in 15 years, so I don't mean to say it's buggy. Your right, it's a matter of what people are used to. Though if I'd played the original DM now, I would say the same thing about it. I think the faster projectile speeds is an improvement. Just a matter of preference I suppose.

[Gambit37]

In the case of DSB, it would be easier for all involved for me to just work with an artsy person to figure out some sensible default coordinates and a field of view, and implement dynamically scaled flooritems, pickableuppables, etc., out to that new larger field of view.

I would be happy to work with you on this. Bearing in mind my comments below!

But RTC develops habits of thought, and dungeon designers are used to working with RTC's interface and around RTC's constraints. To someone accustomed to RTC, DSB seems dauntingly complex and you have to go back through that very first learning curve again

That is basically the problem for me. I've been tinkering with RTC for years. If I'm honest, I simply don't have the time to learn a new tool. I'd have no life if I started to learn DSB!

Soaponarope's sentiments pretty much match mine:

Working on creative aspects is something that is very fun for me; learning Lua is not. I don't know how much I will need to learn exactly as I've only just started playing around with the basics, but from other comments it seems programming experience is needed. I am quite interested in seeing what DSB offers, though on a first look RTC's better visual editing more appeals to me.

I'm exactly the same, and while I have some basic programming skills - even if I had the time to learn DSB - I'm hesitant to get into something which requires that mindset as it's simply not the way I think. I find it a huge challenge thinking in programming logic terms, as well as learning language syntax. Those barriers don't exist with RTC.

From playing the test dungeon it looks like some cool things can be done(and I'm sure many more) and I am glad to see an editor that pushes beyond RTC in it's limits.

There are few awesome things in DSB that instantly make it more attractive than RTC:

1) Default monster AI is far superior. And with custom programming, you can make any monster do anything.
2) Feature organisation: By virtue of including text files, you can keep small features in their own easy to find locations
3) Graphics organisation: being able to specify graphics in multiple organised text files, as well as one big icon sheet, is very cool. Once you start adding many graphics to a project in RTC, its tabbed menu system becomes unusable. I waste huge amounts of time trying to find things in those menus. (The whole "Scalings" taxonomy is just nuts.)
4) Being able to write text directly to a screen is very useful! In RTC, for scene setting stuff, if you want full screen stuff, you have to prepare the whole thing as a bitmap which starts to rapidly increase filesize.

Maybe the only way we can solve this is for people to work in teams: visually creative types can do the basic dungeon layout, basic mech, and any graphics required. A programmer type could then help out with anything complex that the visual person can't do. But of course, as we've proved several times already, everyone seems to prefer working on their own.....

[ian_scho]

2) Feature organisation: By virtue of including text files, you can keep small features in their own easy to find locations

Not only is it kept nice and tidy but sharing your work as well is much easier. Imagine downloading a dungeon and being able to use someone else's monster or magic spell. Very cool.

[Twiggy]

I'm just stumbling onto DSB, but this thread seems to address my main concern, so I'll ask in here

I know absolutely zero about coding/programming. Going through the tutorial pinned in here, the basic interface seems straight forward enough and I think I can handle that and do stuff, but I've seen a few people mentioning now about the coding side of things. My question is, at what stage would I need to start doing that? Like can I make a basic dungeon that I can load up and play with my mum like I can in RTC and coding is just when I want it to start doing more complex thing. Or will I need to start coding to say run multiple triggers/mechanics, or make monsters do stuff and so on.

If you could give me a heads up on that, it'd be fantastic. I understand the main guts of RTC, but I've only worked with that for a couple of months, so if I can run around with DSB, I have no issues switching it up

Twiggy

[Sophia]

can I make a basic dungeon that I can load up and play with my mum like I can in RTC and coding is just when I want it to start doing more complex thing.

This.
For example, if you wanted to duplicate DM or CSB, you wouldn't need to write any custom code at all.

I think the biggest stumbling block is that adding/replacing graphics requires a small amount of code, but it's a one-line thing that can be copied and pasted and is honestly more akin to creating a configuration file.

[Mon Ful Ir]

... and I've created a library of alternative graphics that people can download and work from. There are a few bits of code in there as well, to allow underwater levels and monsters that turn you to stone, but you can disregard those without any problems at all.

[Twiggy]

Ooo. I'll have a look at it later tonight

[Gambit37]

Do you know, I think I had a light_bulb moment regarding DSB yesterday! It wasn't actually too hard to code up a new title screen based on looking at existing code. And I'm begining to see that in fact many things are a lot quicker than in RTC: although writing code seems at first daunting, once you know what you're doing, it's much quicker than working with RTCs weird menus. It's also a no brainer to see how much more control DSB gives -- I mean, in RTC you can't really create proper custom monsters and you can't even create new projectile spells (not properly anyway). DSB will allow *anything* you can imagine, within reason

I also love that I have complete control over code, assets and file organisation and can actually find my stuff easily. Little things like one big sheet for all the icons will make things a lot simpler in future.

I still find ESB hard to use due to its lack of friendly lists and having to type stuff in. That does slow down the editing process significantly. If lists were available instead of all the "type in" fields, I think dungeon editing would be a lot quicker. And I miss not having my item icons show up in the dungeon but I suppose I will get used to that, though it does make editing harder. Still, I coped back in the day with DMute and CSBuild, so I guess I just have to undo my RTC way of thinking. NOt easy, but I'm now more willing to give it a go.

There's a few things that I'd still like to see in DSB: a bigger viewport, antialiasing of distant items and being able to control the distance shading factor would all be nice, but I can live without those for now.

Anyway, now that I'm aware of how it all works and am (almost) happy about coding, I can see bright times ahead for my custom game. It's been withering away on my drive for ages due to bugs and problems in RTC, and I can finally see that it's completable in DSB. Now I just need to find the time!

[Sophia]

being able to control the distance shading factor

For what it's worth, this one is already doable, just not exposed to ESB.

However, you can do dsb_level_tint(level, { R,G,B } )

Code: Select all

-- fade everything to a dark red
dsb_level_tint(0, { 96, 0, 0 } )

The best way is probably to stick it in dungeon.lua yourself. ESB will load and save the tint properly, even though it doesn't expose an interface to edit it. Alternatively, you can place it in sys_game_start or something that will run after the dungeon is already loaded, but it's generally speaking not the best idea to put dungeon configuration there.