WadC Community Project?

[jute]

WadC, available here: http://strlen.com/wadc/ is a severely underused mapping tool. How about a community project to encourage mappers to get to know it a little better? Would anyone be interested?

[MajorRawne]

No offence, but why would I want to spend weeks learning how to program when I can use the same amount of time to draw five maps in Doom Builder?

[40oz]

I'm scared.

[maggot202]

Mirc AND Netscape? this program must be older than Zeus!
In any case, i agree with MajorRawne. What would be the comparative advantages of using WadC over DB?

[EarthQuake]

This is pretty much only useful for precise, yet tedious work. I don't see how the average wad author is going to benefit from it. Besides, the maps would probably turn out boring as hell, copy-pasta and symmetry abound.

[MP2E]

I think WadC is a pretty cool idea but I can see why people are adverse to it, some work is being done on it at this git repo :
http://github.com/jmtd/wadc

[GreyGhost]

This could be fun and more than a little challenging. Before committing myself I'll first have to see if I can -
A - get the program set up and running
B - produce a simple map without losing my sanity

[DeathevokatioN]

Yeah, this actually could be fun, but I'm still new to mapping and lose my sanity trying to make a map through Doom Builder 2 as it is, so I'd probably end up shooting myself if I tried with this program.

Good luck though.

[jute]

I wonder if WadC might have better results in the hands of some of the source-port programmers around here, who are familiar with this kind of work environment, rather than established wad authors used to traditional programs.

[Quasar]

May I suggest that a requirement of the project be that it is open-source. Otherwise how will we know that the maps were created with WadC? ;)

[Scet]

jute said:

I wonder if WadC might have better results in the hands of some of the source-port programmers around here, who are familiar with this kind of work environment, rather than established wad authors used to traditional programs.

I'm a C programmer and even though I don't make maps, I'm pretty sure I'd have an easier time in Doom Builder then using this thing. Even if I understood the commands it would still take much longer than a mouse driven editor. Hell the author of the program couldn't be bothered to make something that wasn't symmetrical and copy-pasted.

Editing would be a nightmare.
"Got to find where sector 541 is declared. Have to find this line. Have to move this vertex by 1 which means changing four functions and checking that doesn't mess anything else up."

There's a reason some tasks are done via text and others graphically.

[Gez]

MAP12 of Crucified Dreams and MAP32 of 10 Sectors were made with WadC.

[GreyGhost]

I've just finished having a play around with WadC and am pleased to say my head didn't explode, though I encountered a couple of major issues -

• Maps (and all objects within them) are rotated 180 degrees when saves as wads
• Map lumps are created out-of-sequence

The second issue's a show-stopper, ZDoom just glumly says "Execution could not continue - 'LINEDEFS' not found in MAP01" until I use a wad editor to manually re-sequence the lumps.

Looks like there's a little java debugging to do before this project can get off the ground.

[Guest DILDOMASTER666]

If I can honestly figure out how to make things in WadC, I'll give it a try.

[GreyGhost]

After a bit if trial & error I've squashed the lump sequence bug and got my little three sector masterpiece to build and run in ZDoom. Here's the patched WadC if anyone wants to try it out. I'll take a look at the 180 degree rotation issue after I've added a few more sectors and things to my map.

Edit - Mapping's fairly tedious compared to DoomBuilder, you need to plan ahead and/or comment your script extensively to aid navigation. I've yet to find a way of setting linedef flags, though maybe I haven't looked hard enough. Here's what I've done so far while learning to use the editor - roughly 180 lines of code that should look familiar.


(1440x900 - 39k)

[Lorenzo]

I tried to use wadC but found it too complicated.

... but it should be nice to have a recursive program which build fractal-like wads.

[kristus]

IIRC from talking to Aard about it some years ago. He never really expected people to use it for mapping. It was more a fun proof of concept that he made.

Though I do believe he did use it when he made his map for Crucified Dreams.

[printz]

Looks like those hardware description languages. Probably it could be useful to design very patterned, repetitive layouts that are tedious for manual painting.

[Afterglow]

kristus said:

Though I do believe Aard did use WadC when he made his map for Crucified Dreams.

He did but I made some texture selection and alignment modifications myself in Doom Builder.

[Karnizero]

For people that knows programming like C/C++/C#/JAVA/etc... it's not so hard to build a map with WadC.

[Jon]

Afterglow said:

He did but I made some texture selection and alignment modifications myself in Doom Builder.

The source to the level in question is bundled with WadC.

I think Aard saw WadC as an interesting experiment and an opportunity to try something new and unique, not as a direct competitor to Doom Builder (or anything else: it preceded Doom Builder by years. It's nearly 10 years old!). He experimented and then moved on.

It's absolutely true that you can create boring, monotonous and repetitive maps with WadC. And there's no point in that. But my perspective on it is fairly different to most people in this thread it seems. To me, the hand-built wads being churned out week after week are boring and repetitive: I've seen it all before, time and time again. Just in a different way.

WadC still interests me because I think it (or the concept it embodies, at least) has a lot of untapped potential. It's possible to create novel stuff using wadc, that you can't with other tools currently around. But nobody is really exploring it.

I haven't looked at it for many months but in 2006 or so I spent a lot of time exploring some concepts with it. I unfortunately lost some of that work since then. But I recall for example, writing some routines to create a length of sewer-like tunnel. I did a bit of work to make routines for a straight bit, a left turn, a right turn, etc.; quite basic. I then used these routines to try out some map layouts. I could vary the precise size of the tunnels and I had several "strata" of tunnels in the map at different heights to each other, all interconnected. I set the "water level" as an argument to the routines and the routines themselves handled boom 242 type sectors so that if the tunnel was below this "water level", it was submerged (with the appropriate palette shift), otherwise, you were e.g. knee deep in it.

The beauty of this was, I could mock up the tunnel routine concept, get the routine API stable, then mock up some layouts, then go back and tweak/refine the tunnel routines, and not have to redo all my layouts. I could put loads of work into detailing the tunnel routines, adding random variation, handling complex light patterns, and have that work take place entirely separate from structuring the map flow.

It was great for abstracting a lot of the work for boom specials. I explored building extremely elaborate conveyor/dummy sequences. I got a hexen-style room light change effect going when you opened a door which was very easy to reuse.

Of course a lot of that sort-of-stuff you could more easily achieve using scripting in another port such as zdoom or with fragglescript.

[GreyGhost]

They said it couldn't be done (actually, no-one's said anything), they said I was crazy (that's probably common knowledge), but after days of boiling brain cells I probably can't spare - IT'S FINISHED - for now.

WadC's default behaviour of deleting mid-textures when there's an upper/lower texture visible means there's no balcony railing or grille for the imp cage, though I could probably work around that. Had some sector-in-sector drawing problems in the North room and wound up taking a few liberties there. Apart from that, it's a fairly accurate reproduction and not bad for an unfinished editor.


(1440x900 - 59k)

[Jon]

GreyGhost said:

They said it couldn't be done (actually, no-one's said anything), they said I was crazy (that's probably common knowledge), but after days of boiling brain cells I probably can't spare - IT'S FINISHED - for now.

I look forward to trying this!

WadC's default behaviour of deleting mid-textures when there's an upper/lower texture visible

I think it just deletes them on all 2s linedefs. I fixed this a while ago, but haven't released a newer version of wadc. The 'midtex' branch in my github repo http://github.com/jmtd/wadc/tree/midtex adds a 'midtex' command to the language which toggles this behaviour.

Other changes I made but haven't bundled up: a command to toggle the boom 'friendly monster' flag on things; a command to toggle whether 2s linedefs should be impassable.

[GreyGhost]

Jon said:

The 'midtex' branch in my github repo http://github.com/jmtd/wadc/tree/midtex adds a 'midtex' command to the language which toggles this behaviour.

And I look forward to trying this. :)

[jute]

I'm very happy to see a dialog going about WadC. I really enjoyed reading about your experiments with it, Jon. I suspect there is a lot of untapped potential for making maps in a totally different way.

[40oz]

Is it fast?

(by fast I mean would a regular 40oz sized map take less than 6 hours?)

[GreyGhost]

Going by my brief experience - mapping in WadC is slower than using Doom Builder. Switching to a text-based editor (where you're basically programming your map) means adapting to a different design methodology where there's no 3D edit mode and simple tasks like flipping linedefs aren't quite so simple. Fortunately, the basic command set's small so there's not a lot to memorise, the most tedious part is probably keeping track of which compass-point you're facing when drawing a linedef. Give it a shot - what have you got to loose apart from your sanity. :-)

[Jon]

Inspired by this thread, I've picked up what I was working on some time ago and hope to create a release soon. Aard gave me permission to take over wadc a while ago as he's done with it but I never put the time in.

Anyway, here's a version 1.2 release candidate, adding the three language keywords I mentioned:

http://www.staff.ncl.ac.uk/jon.dowland/tmp/wadc_1.2-rc1.zip

The 1.2_features.wl file in the examples directory can be used to demo them (untitled.wad in there is the generated WAD file, untitled2.wad is the generated WAD file + glbsp nodes).

GreyGhost said:

After a bit if trial & error I've squashed the lump sequence bug and got my little three sector masterpiece to build and run in ZDoom. Here's the patched WadC if anyone wants to try it out.

I'd like to add this patch to my source tree but mediafire is not working for me. Can you send me (just) the diff?

[GreyGhost]

Mediafire occasionally fails for me as well. Check your PM's.

I'll try out the release candidate tomorrow.

[Worst]

Nice work, I've been using WadC lately, so work on it is great news.

some suggestions,

a way to set the thing tag in hexen mode, setthinghexen(type,arg1,arg2,arg3,arg4,arg5) is missing it atm.

Also a possibility to set the line activation type in hexen would be welcome too I guess.

GreyGhost said:

..the most tedious part is probably keeping track of which compass-point you're facing when drawing a linedef.

You can avoid that, if you just use step(x,y) for drawing the lines in the imaginary grid.. a lot easier than trying to remember the orientation :P