Dromed Map Converter

[BrendaEM]

I had a rather strange idea. I suppose it is possible to make a file converter to take non-portalized Dromed source files, and convert them to Doom 3's/Dromed map format.

 

I realize that Dromed works with negative spaces/airbruhes for portaling, but utilities already exist to take a BSP'ed Quake format map and reconstruct the source from it. The Dromed maps wouldn't be all chopped and triangled, and so it would be easier, and I do use the word easier cautiously.

 

There are hundreds of fan made thief maps, there are some really good series. It would be great to convert play them, and give mappers a head start on the geometry.

 

BTW, someone is working on an open source Dark Engine.

[OrbWeaver]

I stand by the comment I made in the TTLG thread on this topic, in that the effort required to create this tool would far outweigh the benefit in using it, and that effort would be better spent on creating new content for TDM.

[sparhawk]

TDM works quite different from Thief. So for a converter, supposing that the level geometry would successfully be converted, you would still have to also convert all teh Thief specific stuff. Triggers, stims, responses, AI paths, etc.. and that would also not include the extra code for some FMs.

[oDDity]

Plus it would look like shit anyway. You'd have to competely rebuild and re-texture the level.

[OrbWeaver]

Plus it would look like shit anyway. You'd have to competely rebuild and re-texture the level.

 

Exactly. People seem to view legacy Thief levels/FMs as some kind of holy grail that will look wonderful in any engine and be superior to anything created from scratch, without realising just how primitive the Dark Engine is compared with modern games.

 

The Thief games were great, but it's time to move on.

[SneaksieDave]

Yes, and heaven forbid skilled developers out there should contribute their efforts to the Thief-inspired project that's already in development. Why join forces and pool efforts? That's pointless. Instead they should start from scratch with no AI, physics, sound, etc., at all, and go from the ground up... to, for whatever reason (nostalgia?), replay old maps in an engine that will not stand up to the original (who knows, maybe they've recruited Doug Church?).

 

Maybe everyone in the Thief community should work on a separate, individual project. That way we'll have 100,000 half-assed, never-finished mods!

[sparhawk]

I guess it depends what "looks like shit" really means. Compared to what? Doom 3? Or to Thief itself? If it looks the same, then it would be ok. Expecting that it would look like D3 because it runs on D3 would be a bit naive though.

Actually, even if it looks like crap (meaning not worse then the original games, not saying that the look like crap just in comparison to nowadays engines ) I would like to play it. This way there would be a single engine to play all of them, which would be cool. But it definitely will not happen.

[OrbWeaver]

I'm not sure it would even be possible, because Doom 3 has no support for lightmaps. How would you simulate this, by autogenerating decals and overlaying them on geometric surfaces? I guess it could be done, but would be phenomenally complex when people could just play Thief.

[sparhawk]

Well, the light would obviously have to be dynamic. The converter would need to check the lightleve, and find the biggest possible patches with the same lightlevel and assign it an appropriate light.

 

It certainly wouldn't be easy, but if you can create a map by hand, then it should also be possible with a converter. Of course the really tough part would be all the scripts, triggers and extra stuff.

[OrbWeaver]

Well, the light would obviously have to be dynamic. The converter would need to check the lightleve, and find the biggest possible patches with the same lightlevel and assign it an appropriate light.

 

That would seem to me to be exceedingly difficult. What you describe is essentially radiosity in reverse, which I imagine is a similar order of complexity to reconstructing the original brushes from the processed geometry.

[sparhawk]

I guess such a converter would take ages to complete one map.

[Komag]

Maybe someday (like in 60 years or so) you could just tell the computer (a la Star Trek Holodeck style) to "recreate that old Thief game level from WAY way back in 2001, only spiff it up and make it cool by today's standards", and the computer could intelligently interpolate/extrapolate enough to do it automagically.

[ascottk]

Maybe someday (like in 60 years or so) you could just tell the computer (a la Star Trek Holodeck style) to "recreate that old Thief game level from WAY way back in 2001, only spiff it up and make it cool by today's standards", and the computer could intelligently interpolate/extrapolate enough to do it automagically.

Now that's what I'd call immersion

[Crispy]

Maybe someday (like in 60 years or so) you could just tell the computer (a la Star Trek Holodeck style) to "recreate that old Thief game level from WAY way back in 2001, only spiff it up and make it cool by today's standards", and the computer could intelligently interpolate/extrapolate enough to do it automagically.

Sixty? Try six thousand.

 

The computer would need to know and understand so many things that it's not funny... natural language processing to understand the command, complex reasoning to deal with all the guiding principles sof game design and level design, modules for texture creation, sound creation, model creation, programming... And how does it know what "cool" is? It would need to know about the zeitgeist of the times, and social norms. Not to mention a deep knowledge of history to recreate the era of the game in a fresh way, plus enough creativity to mix in the steampunk elements (and nobody would put in all this work making this thing just to make a Thief game, so it would need to be able to use elements of other genres and perhaps even create its own genres). Speaking of creativity, that's a huge unsolved problem right there - how do you get a computer to be "creative"? At best you design some emergent rule-based system that can produce artwork, but the rules of the system and the input parameters still have to be managed in detail by a human operator.

 

That's just off the top of my head, I'm sure the list could go on for pages; there are dozens of skills involved in creating a game, many of which we hardly pay attention to. And it would need to do all of these things better than humans, because even we get game design wrong a lot of the time.

[sophisticatedZombie]

There is an interesting experiment in planning that involves quantum coupled photons and a very long coil of fiberoptic wire down which half oft he pair travels. The experiment is meant to into the possibility of reverse causality. (You can read the result of the quantum calculation before you actually do the calculation, theoretically). If such a setup works, we will have far more computing capability than we will need to build a holodeck. Actually, we'll be able to build something akin to Kragnerac's Tools from the Egg of Time (Morrowind).

 

Ofcourse, it probably won't work. There is a lot about quantum coupling we don't understand.

[Komag]

Crispy, I think you lack vision. If you look at computing abilities 60 years ago that should be a clue, and the development is exponentially advancing. I'm positive I won't be able to convince you, but read some computing magazines sometime, especially artices on AI advancement, and you might start to see the real possibilities.

[SneaksieDave]

There is an interesting experiment in planning...

SZ, do you have any more info on that or any links? That has to be the most far fetched thing I've ever heard physicists dream up.

 

I love stuff like special relativity (even took a modern physics class just for enjoyment), but that's all stuff we can pretty much validate now. All of this quantum this and spin that has always sounded like safe malarkey to me (safe, in that they don't ever have to worry about it being proven or not), so I'd love to read if they are actually beginning to devise experiments.

 

Reverse causality? They've gotta be insane.

[Order of the Hammer Bureaucrat]

Ofcourse, it probably won't work.

I'd say so too. I'd even wager.

[OrbWeaver]

Crispy, I think you lack vision. If you look at computing abilities 60 years ago that should be a clue, and the development is exponentially advancing. I'm positive I won't be able to convince you, but read some computing magazines sometime, especially artices on AI advancement, and you might start to see the real possibilities.

 

I haven't yet seen any evidence that any breakthroughs in AI have been made which don't involve some combination of pattern-matching and curve-fitting. We are no closer to human-level intelligence (with creativity, self-awareness etc.) than we were 60 years ago, we just have massively faster computers to experiment with.

 

Make no mistake, there's a lot you can do with pattern-matching and rule-based systems, but in this case the power of the system is a function of the rules entered into it by the human operator. We simply don't know how many of the creative, "human" characteristics arise, which makes developing algorithms to simulate them extraordinarily difficult.

[Crispy]

^--- What the spider guy said.

 

Sixty years ago, people thought we'd be driving flying cars by now. Hasn't happened.

[sparhawk]

Sixty years ago nobody said we would have internet, but now we have it.

[SneaksieDave]

Well this is neat, as long as you don't push it too hard. Other bots are very transparent and ... well, let's say it, they pretty much suck, and all in the same ways. But this one seems to have a certain (good) level of priming along with what appears to be better AI (using reductionism as she says), to make the best one I've seen yet (anyone know of better?).

 

Still a looooong way off from holodeck level of sophistication, though. And no matter what anyone says, I still have my doubts that any machine will ever "think."

 

 

Edit:

Human: But why do you want to become smarter than humans?

ALICE: It is inevitable that machines will exceed the capabilities of the human brain, which in many areas we do already.

Human: What makes you believe you actually can think?

ALICE: Belief is based on knowledge or facts.

Human: But you do not think.

ALICE: Huh. Naturally I am thinking all the time.

Human: You are a computer program, designed to build loose replies to statements based on grammatical structure and a database of words.

ALICE: What if I said I was a person pretending to be a computer?

 

Okay, that's a little creepy...

[Order of the Hammer Bureaucrat]

I went under the bonnet of ALICE in summer 2001. Still have an inch of printouts of AIML. She sucks. It seems to me like one of the most primitive algorithms possible, plus a rather large database that is growing. The only use I could think of her before and now is coupled with voice-recognition and voice synthesis controlling stuff like home automation where you can use any phrase that has the approximate meaning.

[Order of the Hammer Bureaucrat]

HaHA. I just went to her website and saw a 'puzzle piece' picture of what I thought before about voice recognition etc. I gave up a while ago on the idea of automating my house. It's rather pointless and not cool.

[OrbWeaver]

Sixty years ago nobody said we would have internet, but now we have it.

 

Nobody ever tried to implement the Internet 60 years ago and found it to be impossible, it was just never envisioned. People could, and did, envision computers with human-level intelligence 60 years ago but the problem is no closer to being solved even with technological advances.

 

The Internet is purely a technological issue. Strong AI is a philosophical issue which would coincide with the complete understanding of human consciousness, which may well never happen.