c00nsolid - a new Quake3 map editor

[mar]

Hi all,
I'd like to share a simple project I've been working on recently. Pre-alpha stage so take it easy.
The link is here: http://www.mediafire.com/?2uy51zhmdyzlydn
Any feedback would be welcome.
Thanks and enjoy.

mar

[cityy]

What is your plan for this project, how does it differ from GTKradiant and lastly, screenshots?

[CZghost]

The only differences are:
#1 The controls are much hard to get in
#2 There are no preferences at all
#3 Textures doesn't show up

I miss the point of the editor: title is a bit confusing, looks like GTK Radiant after an alcohol party...

[Eraser]

Reminds me of this Quake 1 level editor I ran into recently. There's an interesting video demoing the editor on their site:
http://kristianduske.com/trenchbroom/

[mar]

Thanks for the feedback. It seems it was a bit premature to "release" an unfinished version
As for the differences (quoting readme), despite the fact the there is zero code overlap:
(replace solids with brushes if you want)
- works with multiple maps at the same time
- vertex/edge editing computes convex hull instead of collapsing the solids when you move too far
- poly builder: draw arbitrary concave polygon in 2d view and turn it into solids
- CSG intersection
- center 2d views on selection
- true "round" cylinder/cone patches
As for the screenshots: I don't know how to post them here without having to use external uploader like imageshack.

CZghost's point 3 isn't valid - one has to decompress pkx files and run the editor from that directory. I know it's clumsy
but as I said it's unfinished yet, I just wanted some feedback.
As for the point: it's simply my code I enjoy to work on, I could as well ask you what's the point of creating maps, but I get it
As for the name: it's not that important to me however weird it may sound.

Yes I have seen Trenchbroom. I like it a lot, it's really different from anything else out there.
It's simple yet very well done. I will play with it a bit more (never understood how to switch planes/axes of movement in 3d), I also liked many cool ideas and brush outlines/grid.
I also tried Quark but it looked too complicated to me, not sure. Also the software renderer looked weird IMHO.
The Radiant way (simplicity) looked the way to go.
I never liked those handles and random colors in Valve's WorldCraft/Hammer.

My only problem with Radiant was version 1.5, it was simply full of bugs. No wonder they went back to 1.4 and based 1.6 on that.
I guess noone used 1.5 for serious development anyway but I may be wrong. DarkRadiant fork looks cool but I guess that's for DarkMod only.

[obsidian]

I understand that this is your personal project and that you're doing it as a hobby. But have you thought about contributing to the GtkRadiant project? If you have ideas on how to improve an editor, that seems like the best way to go. On the other hand, if this is just about practicing your coding skills, all the power to you and all the best.

[mar]

Well I had a quick look at Radiant source (1.5) and no - I don't want contribute. I have to say the code is a complicated mess, unfortunately.
Obviously not written by Carmack himself who writes simple, clean and efficient code (that's what makes him a genius among other things).
My code needs lots of refactoring too, but I can handle it easily so far. Plus I believe that having based it on Qt instead of Gtk/wxWidgets was the right decision.
I have ideas about what needs to be done, the todo list is quite large. I'd like automatic patch capping for example.
I'm also thinking about extrude faces option but hard to say if it would be of any use.
I believe I can make something competitive but there's still a lot of work ahead.
I'm not a mapper so I can't distinguish between which features are useful and which are useless.
I consider Radiant's make hollow useless for example (in its current state), I can imagine a much better solution (with clean brush topology).
Originally I wrote it for my engine which can build map files. Then I thought it may be used for quake mapping as well and that it may be useful to others, but obviously not in its current state, I'm well aware of that.
Hey and thanks for the suggestions.

[obsidian]

Wrong version, GtkRadiant 1.5 was a very different branch that decided to do some very different stuff under the eyes of a different set of developers.

The GtkRadiant 1.6 project is very active and is based on the older 1.4 branch. It is headed by TTimo, formerly of id Software who oversaw most of the development for 1.4 and now 1.6.

Info is here:
http://icculus.org/gtkradiant/

Source is here:
https://github.com/TTimo/GtkRadiant/

[Bliccer]

Of course creating an own project is always a huge motivation at the beginning, but in the end I guess it will be more frustrating to see that no one is using it. So helping in the Radiant project, I guess, would be more satisfying. Afaik there is a thread somewhere around here with pretty good ideas for improving this splendid editor (which is imo the only editor I will ever use for q3mapping, cause it's just so fast to handle). Adding your ideas such as the simultaneous opening of maps would be premium.
For me personally, there would be no sense in creating an editor from scratch for such and old engine, when there is already an well established editor out there.
The DarkEditor is for the DarkMod only based on the Q4/D3 Radiant.
Just my two cents...

[mar]

Thanks for the clarification. I thought 1.5 was an official version. I was shocked by the amount of bugs which rendered it almost useless IMHO. Too many cooks spoiled the broth it seems.
Starting from 1.4 seems like a wise decision.
I think that Bliccer is right and noone will probably use my editor anyway.
Quake3 engine is old but it's still my all-time favorite.
I'm a bit stubborn and will continue my editor anyway. Creating a map in your own editor, even if both are crap is fun

[Eraser]

I'd say keep on going with it. If you don't feel like working on GtkRadiant 1.6, then that's fine and I know how much fun it can be working on a project that's your own and that you're proud of. Things you do don't have to be useful in the end.

[mar]

Thanks Eraser. I will try to dig up the thread Bliccer mentioned to hopefully scavenge some ideas
I have another idea, rendering edges that share one or more faces mapped with caulk/other common textures semi-transparent.
A properly caulked map should then look better in 2D because otherwise it can be a mess even with clean brushwork.

[themuffinman]

Agreed on the disaster that was Radiant 1.5.0 and these guys make sense - a new editor is really unlikely to attract much attention at this point. However, if you could offer something unique and incredibly useful then you might make a success out of it.

An idea I've had for a while is to create a completely in-game level editor, build right into idTech3. It would give the ability to simultaneously create and test the level and possibly even host 'level design servers' for multiple mappers to work on a map at once. I think it could be doable. Item entities could definitely be easily moved about and placed in different spots and assigned different keys in realtime. Recent ioquake3 revisions have a GLSL support which still supports traditional Q3 shaders so realtime lighting of levels could come in handy for getting a feel of what the lighting may look like after compiling. The XReal engine mod can load .map files so the source code could help too. Orthographic views and the overall interface could be quite a hassle to get right though, etc.. It would be a pretty big undertaking but I think the results would be worth it.

[mar]

There is already an engine that can do cooperative ingame editing: Cube2 Sauerbraten http://sauerbraten.org/ (unless you already know)
It's awesome because it's really very different from usual bsp based engines and thus very original. I like it a lot.