RTC vs DSB

[Gambit37]

[Admin Note: Split from this RTCWM thread: http://dungeon-master.com/forum/viewtop ... 32&t=25048]
Topic then moved to custom dungeon forum due to its general nature.

PLEASE DO NOT LET THIS THREAD BECOME A FLAME WAR OR IT WILL BE LOCKED AND MOVED! We collectively love all the creative endeavours here! - b.


What's the reason you want to use RTC over DSB, given that RTC is no longer updated and is much less flexible?

[ebeneezergude]

Thanks Gambit, no real reason really, just that RTC editor seemed more accessible to use, I guess. ESB displays generic icons for each object, whereas RTC editor allows you to tell the difference between wall text and a lever, or a cheese and a scroll. I found this very intuitive. ESB, given that it is slightly easier to use in terms of editting simple corridor layouts and placement of objects, is slightly baffling why the in-editor icons are mostly generic rather than specific. That would seal the deal for me. Willing to give it another go though!

EDIT: and RTC seemed more accessible for more difficult things like creating new objects.

[Gambit37]

Presumably someone has managed to 'translate' all the required co-ords for each bitmap type to ensure RTCWM to RTC is correctly registered in the dungeon.txt?

Nope, not as far as I know. You're treading new ground.

The whole thing is trial and error and this is why no-one ever previously documented a method for it as it's too bloody time consuming...... firing up RTC, testing a position, quitting RTC, adjusting in editor, firing up RTC, etc.... it's just a crazy, crazy waste of anyone's time. DSB starts up almost instantly and makes all this much simpler.

I understand your reservations over a lack of visual editing in ESB (I had them too), but all the other customisation possibilities offered by DSB ultimately make it worth the pain of a less flashy visual editor.

[Paul Stevens]

customisation possibilities offered by DSB ultimately make it worth the pain

This is from a man who, for years, declared that
he was totally incapable of writing a line of code.

[Gambit37]

Not sure I ever said I was totally incapable? If I did, then that was incorrect. I'd be a pretty poor web-designer if I couldn't write a line of code, since understanding code is a core part of the job!

My frustration has always been that I don't find programming especially "intuitive": I can cope with simple things, but grand conceptual structures in sophisticated programming (and the sort of abstract thinking that real programmers are good at) both confuse and bore me. Programming for me is just a means to an end in the process of creativity, so I only learn the minimum I need to do what I want to achieve. I prefer visual tools over programming as I find them easier to learn and master. I also didn't really understand the innards of DSB when I first started, which put me off.

But to be honest, I think it was more about not wanting to lose several years of tinkering in RTC to having to start over in a new engine. But I'm glad I did

[Seriously Unserious]

I'm also in the camp of switching over to DSB when I've finished my current series of custom dungeons. I would switch now but I've done so much work in RTC that I don't want to lose to a new engine so I'll stick with it for now so I can use what I've already got but within the next year or so I should be ready to make the switch for other unrelated dungeons (to my Lynchgate series).

[ebeneezergude]

Well now I guess I'll have to have another tinker with DSB....!!! I got the graphics imported from RTCWM already (simple file drop and name change as per DSB tutorials), and can use the editor very easily as it's very well written (although my icons gripe is still there - could Sophia program this in to allow users to create icons for objects...? I see a cells.pcx file....), however I haven't yet got my head around how to create lua files for new things, how to structure the lua files, etc. Will have to give this a crack.

A more general query, only partially related to RTCWM but more about the decision for which engine to go with: If, assuming I do a dungeon using RTC using the standard wallset, with no custom walls, but with additional self-made objects/monsters/decorations, and, assuming I save several times a day, is RTC good enough as an engine to go with? The editor is very capable and user friendly in terms of creating new monsters and objects, for example. As a comparison to the original Amiga/ST engines, how does RTC stack up in terms of accurately recreating the original mechanics, feel, authenticity, etc?

And how about DSB for that matter? How does that stack up? I read on another thread glowing praise for DSB calling it the 'gem' of clones.

I've read both author's statements on this question, where they state that subtle differences are possible between their engine and the original, but was interested to hear your views.

Have heard that RTC can have some bad crashes, and is not actively supported, but is it useable, if handled with care, enough to commit to a medium to large project? Are there any catestrophic issues that cannot be worked around using diligent file saving?

I'm now floundering with indecision... best tinker some more with dsb

[Seriously Unserious]

If you've already done a bunch of work using RTC then I'd recommend you stick with it, at least for any custom dungeons that would draw from that work. RTC may not be actively under development any more, but it's still a heavily used DM clone engine and can do some amazing things. One difference I've heard about between RTC and DSB is the AI for monster behavior is better in DSB, monsters act in a more intelligent way in DSB, whereas in RTC they seem to behave closer to the original DM patterns. RTC still does have noticeable differences, but that's mostly because Geroge Gilbert didn't have access to the original DM design team or the monster behavior algorithms designed by the DM development team.

Sofia, the developer of DSB, seems to have desired to upgrade the monster behavior model, rather then attempt to recreate the original.

Also, DSB uses the Lua scripting language, so if you feel comfortable with working with the code, then DSB is a viable option, otherwise I'd recommend sticking with RTC and it's more developed graphical editor, where scripting code is totally unnecessary. I'm not familiar with Lua myself, but if it's anything like scripting languages like PHP, Javascript, Perl, etc, then it should allow for more versatile decision making and looping controls then the RTC relays can provide, and that's one of the things that I find attractive about trying DSB.

I'm just waiting until I've finished the projects I've already started in RTC and DSB and ESB are a bit further along in their development.

As far as the bugs remaining in RTC, I haven't had any major issues with any of them so far, but I do keep back-ups of my dungeon file so if something does go wrong I limit the amount of work lost due to, eg, a corrupted mydungeon.txt file or anything like that. Which is good advice no matter what tool you're using anyways.

[Gambit37]

Take a look through the RTC bugs forum to see what's there:
http://dungeon-master.com/forum/viewforum.php?f=4

As long as you don't want to do any of the stuff that's buggy, you should be fine to continue using RTC.

I've nothing against RTC -- I spent at least 5 years of my spare time helping debug it and make suggestions for improvement! It's a flexible and creative tool that allows you to build some very cool things. Unfortunately for me though I got to a point where I was pushing it in ways it simply couldn't handle which is why I switched to DSB.

There were a lot of things that made me switch, but the four main things were

1) Huge limitations in the creation of custom spells http://dungeon-master.com/forum/viewtop ... =6&t=25910
2) Adding hundreds of new graphics and working out their offsets using the visual importer -- boring, too time consuming, inaccurate.
3) Issues with sounds not stopping on player death http://dungeon-master.com/forum/viewtop ... =4&t=27877
4) Slow boot up times -- for testing large graphics heavy dungeons, this was the killer. RTC took a good 20 secs to start.....

vs DSB for the same issues:

1) You can create any kind of spell to do anything you want
2) Graphics mostly "just work", ie, as long as you set them up in your Lua files correctly, the engine does a great job of automatically positioning them correctly.
3) No such sound problems, and a much more flexible sound engine.
4) Almost instant loading (with the default graphics -- does slow down a lot once you start adding your own resources though.)

I've been properly building in DSB for the last year, and while it took some time to understand it (and there's still tons I don't understand), I'm very happy with the results I'm getting, way more so than RTC.

[Sophia]

The boring yet truthful answer as to why I didn't use build in the ability to use icons for objects is simply that my development priorities were elsewhere and I didn't get around to this yet. It's not something I'm hugely excited about doing, however, nor is it something that I am hugely opposed to doing or think would be a terrible task to do or anything like that.

ESB's interface is rather small. I just wonder if the icons wouldn't be too small to be of any use, anyway.

[ebeneezergude]

Hi Sophia, thanks for taking time to reply. It would be brilliant if in the editor you could see what each object was, even if only by some form of type (sword, axe, food, scroll, lever, wall text, etc).

I was wondering if you could set it up so that each 'standard' object (ie base default DM/CSB) calls for a portion of bitmap from a new, enlarged 'cells.pcx' file. In the this file, you as author would not necessarily need to draw each object icon, or icon type (sword, axe, etc), but could let the community know how the cells.pcx file is read (through bitmaps pixels coordinates) so that over time we could populate it with the correct graphics for each icon - ie, saving you the time to draw them yourself?

If a new object/monster is created through lua, then could this be separately associated with a custom icon bitmap for the editor, should the dungeon author/lua programmer wish it.

Is ESB interface easily adaptable to make tiles bigger and able to display the object icons more clearly?

Sorry for asking questions which might be annoying, difficult, or just not even possible, I don't know anything about coding ... Just letting you know my thoughts on the otherwise excellent DSB...me being a programming idiot...

[Sophia]

Is ESB interface easily adaptable to make tiles bigger and able to display the object icons more clearly?

It absolutely is.

ESB gets its icon size from cells.pcx. If you enlarge the size of cells.pcx, then ESB will adapt to the change.
(Double zoom simply enlarges the image internally and the rest "just works" because of this)

I'll look into importing the icon sheet from DM, which should allow for making things look a whole lot nicer.

One of my biggest problems was an aversion to putting ESB metadata into objects.lua. As time has gone on, I've pretty much gotten over this aversion.

[Seriously Unserious]

If you enlarge the size of cells.pcx, then ESB will adapt to the change.
(Double zoom simply enlarges the image internally and the rest "just works" because of this)

Having a gui editor with a zoom feature would be great, especially when working with a large dungeon level, which, judging by the way you've implemented this should be easy to do if you haven't done it already, or to upgrade it to include more levels of zooming, or even a user defined zoom option. That's one major feature RTC Editor is missing, as it gets hard to target objects, telleporter destinations, etc in very large dungeon levels.

[Gambit37]

ESB already has the double zoom feature. If you need to go the other way and zoom out, you wouldn't really be able to see much. But no problem, it has scroll bars for moving around large levels. Why RTC didn't implement that is beyond me

[ebeneezergude]

Sophia, if you could get that working that would be awesome.

[Sophia]

Why RTC didn't implement that is beyond me

It was actually really annoying to implement those, and, had I not made the somewhat impulsive and irrelevant (at the time) decision to make the 'dungeon view' its own window class, it would've been about three times as hard. I'm guessing George did it a different way and just didn't feel like bothering.

The only reason I ended up doing it was because it was something I hated in RTC.

[ebeneezergude]

The RTC I've downloaded has scroll bars and a comprehensive zoom.

[Sophia]

The RTC I've downloaded has scroll bars and a comprehensive zoom.

We were talking about the object picker and teleport destination windows. (At least, I was. I'm pretty sure Gambit was also)

[Gambit37]

Yep, exactly.

[Chaos-Shaman]

having played with RTC for some time, you can now give a monster almost any action as AI. they won't fight each other, you can command them to do anything. follow the party throughout the dungeon if desired. RTC AI is fantastic to say the least. HEY GG, if you still roam this place, have a look at how smart the monster AI is now. everything you wanted is possible.

[Seriously Unserious]

I find one of the greatest strengths of RTC is the fact that it's a nearly fully developed engine, so much of what you need to do can be done. The one main knock against RTC IMO is that there is no way to get a trigger or other object to call up a scripting language that would enable much more complex events then the relays can handle, at least not without doing a lot of extra work and tinkering around with long chains of relays. The main thing is how limited the RTC relays are in condition testing, since each relay can only test up to 2 conditions against each other, so when more complex testing is needed, it gets difficult to get the testing just right to produce he desired result. Being able to call on a script where an if statement could be used to test as many conditions as you want would be very useful in simplifying a lot of dungeon designs.

That is my mind is the main strength of DSB -- its support of a scripting language which can be used to create more versatility, and with less confusing chains of relays. My main knock against DSB at this time is that the GUI editor needs a lot more work to make it more user friendly. DBS is still very much a work in progress but I think is has a great deal of potential and could create things that are difficult to do in RTC.

That said, me and Chaos Shaman have been finding rather creative ways around some of the shortcomings of RTC to create some amazing and complex events.

[Chaos-Shaman]

suspiciously suspicious team work there, hehehe. RTC has slightly been unlocked, there is more to come

[Seriously Unserious]

True.

IMO the biggest problem is the lack of documentation about how to use all the features. Unfortunately, this is a common problem with independent software developers who just don't have access to professional technical writers who's sole purpose is to write documentation like the big developers do.

Sadly, this problem is not fixable with RTC as its developer is no longer supporting it, so any features he didn't explain to anyone we have no way of knowing for sure what they are meant to do. The best we can do is experiment by trial-and-error and try our best to guess at what they're supposed to do.

One drawback with DSB is that to get the most complex actions to happen, and truly get the most out of it requires at least some familiarity with programing, whereas RTC is fully GUI in dungeon design, meaning anyone can design dungeons, without the need to learn how to program first.

I took a look at the DSB wiki pages and it looks like DSB can already do quite a lot. I'm actually looking forward to working with both engines down the road, but I feel DSB still has a year or 2 before it really begins to realize it's potential.

Hopefully, with Sophia still actively working on DSB, the documentation issue can be resolved, with some good instructions on what each feature is, how it's used, and what it does. In the meanwhile, I'll continue to develop my RTC dungeons.

[Gambit37]

IMO the biggest problem is the lack of documentation about how to use all the features. Unfortunately, this is a common problem with independent software developers who just don't have access to professional technical writers who's sole purpose is to write documentation like the big developers do.

It's a pain in the bum, sure, but you have to remember that most of these sorts of projects come about because the developer has "an itch to scratch". In this case, the itch that needed scratching for GG was simply to make a DM clone. All the other stuff came later and I don't blame him for not writing any docs - it's a huge pain in the arse if you've already "scratched your itch", so to speak. The forum was essentially the docs for RTC and it's only through recent work by some amazing forum members that some of it is now available on the Wiki.

The best we can do is experiment by trial-and-error and try our best to guess at what they're supposed to do.

Well, yes, that's exactly how all of us learned it.... and then shared our findings

One drawback with DSB is that to get the most complex actions to happen, and truly get the most out of it requires at least some familiarity with programing, whereas RTC is fully GUI in dungeon design, meaning anyone can design dungeons, without the need to learn how to program first.

I'd agree with this overall, but once you start building complex systems in RTC, you're effectively programming anyway - in a gui interface that's quite clunky, slow and unintuitive for that sort of thing. I shudder at some of the weird combinations of things I had to do in RTC to achieve complex effects, when in DSB it's just one or two lines of fairly straightforward code.

I feel DSB still has a year or 2 before it really begins to realize it's potential.

Err, what? LOL! It'll only realize its potential when people start building dungeons with it! It's more than capable of realizing huge potential right now (way more than RTC ever could), so to say it's TWO YEARS away from that potential (especially when you've not even used it) is both uninformed, and completely unfair. Yes, the editor is not as easy to use as the RTC editor in many instances (in fact, ESB is actually better than RTCEditor for many things), but it's still more than capable of building most things without requiring any programming skills. Hell, I built my first RTC dungeon by hand in a text file, because it didn't even HAVE an editor until v0.26.

Hopefully, with Sophia still actively working on DSB, the documentation issue can be resolved, with some good instructions on what each feature is, how it's used, and what it does. In the meanwhile, I'll continue to develop my RTC dungeons.

You're missing the point. Projects like this won't ever get fully documented by the author, due to the "itch is now scratched" issue I mentioned above. These projects NEED all of us to contribute documentation where we can. The Wiki is open and can be edited by anyone, so instead of complaining about the lack of documentation, perhaps you could start to help fix it?

[Seriously Unserious]

Actually, I'd be happy to contribute Documentation to DSB and ESB. Hopefuly Sophia would have the patience to answer the flood of questions I'd need to ask to find out what features do what of the ones that haven't yet been documented. That would actually be a great way for me to gain experience in this which is part of the IT work I'm looking to get started in anyways.

So if Sophia is interested, I'm willing to volunteer to write detailed documentation on using both DSB and ESB. Of course, I'd have to practice using them to write proper documentation. So Sophia if you're interested in my help in documenting your engine just let me know.

[Sophia]

I feel DSB still has a year or 2 before it really begins to realize it's potential

If you just say stuff like this, I'll simply shrug my shoulders and carry on.
If you tell me concisely what you are missing from DSB, then there is a chance I'll implement it.
And if you dive in and start building something in DSB, you might be able to implement it yourself.

But anyway. I need all the help I can get, especially when it comes to documentation, so consider the offer accepted.

[Chaos-Shaman]

heh, too much to quote there. i mentioned three years ago that it would be the same people using DSB, because as said above, it's the gui. without it, it'll be hard to convince perfectly capable people to create fantasies without messing with the code. i personally look at it and barf. i've no interest in looking at numbers, but that gui, i could spend dayz with it, having fun. i understand why GG did what he did, totally. RTCs main problem was monster AI, according to the messages in the forum. believe me, the problem was so simple to fix, probably why nobody could see it. it is what happens when people put expectations in their minds and shut of the possibilities. i've read so many complaints about RTC that it has effectively pushed the skilled, the artistic away from here. i hope all you guys see that.
making friends here to work with RTC is impossible. i have personal accounts of that, even told by Sophia to quote " do not ask me any questions about RTC. this goes as far back as 2006. that's 7 years of no help for me. thanks
anyhow, considering that gambit thinks i'm from another planet, heh, maybe i am

we're all grown up here now, let's cooperate. please don't bite my head off again gambit, and you shouldn't underestimate people who are trying to belong to this great community. i'm not the kind to go away, time will tell.

[Gambit37]

it's the gui. without it, it'll be hard to convince perfectly capable people to create fantasies without messing with the code. i personally look at it and barf. i've no interest in looking at numbers, but that gui, i could spend dayz with it, having fun.

You'll know that I felt the same way for ages. I just couldn't get my head around building things the DSB way. But we can all change, right? I took the time to investigate DSB and very quickly understood why it's more powerful than RTC. That's not even a personal opinion: when you compare what the two engines can do, feature against feature, DSB wins hands down. As I've said before, I agree that the RTC editor has a nicer GUI, but actually it's so clunky in so many ways that once you start developing in code (in DSB), the speed with which you can setup graphics, new monsters, items, etc far surpasses the workflow you can get with RTC.

RTCs main problem was monster AI, according to the messages in the forum. believe me, the problem was so simple to fix, probably why nobody could see it. it is what happens when people put expectations in their minds and shut of the possibilities.

Cool, well if you've fixed all those issues in RTC, it would be great if you could share how you did it.

i've read so many complaints about RTC that it has effectively pushed the skilled, the artistic away from here.

There are still plenty of skilled, artistic people here doing their own thing; they may have plenty of reasons why they no longer use RTC.

making friends here to work with RTC is impossible.

I disagree. Some individuals may not use it or may have chosen to ignore it for their own reasons, but there are plenty of other forum members who still build in RTC and share their knowledge.

please don't bite my head off again gambit, and you shouldn't underestimate people who are trying to belong to this great community. i'm not the kind to go away, time will tell.

I don't underestimate people, I simply say what I see. From what I understand, you have the view that RTC has been unfairly maligned and you choose to defend that. Fair enough, that's your decision. If I've misunderstood that, please correct me.

RTC is a piece of software that brings out creativity in people. People get joy, fun, excitements out of it. That's fantastic! But eventually, they may decide it no longer suits them and move on. It's no big deal one way or the other, and it's neither your place nor mine to make judgements about those who choose to reject it.

[ebeneezergude]

RTCs main problem was monster AI, according to the messages in the forum. believe me, the problem was so simple to fix, probably why nobody could see it.

Hi Shaman, so how is it fixed? Or is "improved" a better term? Was it broken?

Thanks.

[Sophia]

George once observed that the system of actuators in RTC is Turing complete. What this means is that RTC actuators essentially are a programmable computer. Creating complex mechanics in RTC essentially is programming. There really is no difference-- except RTC's "language" is extremely low level and has some odd conventions related to being tied to a DM dungeon. On the other hand, Lua, used in DSB, has a somewhat steeper learning curve and can perhaps appear imposing, but basic techniques of higher-level programming like variables and procedures are readily apparent, rather than constantly having to be faked through various means. I feel this actually makes complex tasks easier in the long run. As a (somewhat extreme) example, the "Sokoban" puzzle in Surgical Strike was such a complex mess of triggers I actually had to write a separate program to automatically spit out the RTC mechanics in script form, which I then pasted into my dungeon. Sort of a "compiler," as it were. Preventing situations like this was one of my design goals for DSB.