GTKradiant 1.4 source code

[Delirium]

Does anyone please have a link to an SVN or download of any sort?
My google searches were dry so many times :<


any help appreciated

[obsidian]

You probably won't find the 1.4 source code anywhere, but 1.5 and 1.6 (based off of 1.4) is available:

http://qeradiant.com/cgi-bin/trac.cgi/b ... GtkRadiant

There's been some recent activity on the mailinglist and IRC with some interesting code changes.

[VolumetricSteve]

Is there a reason you're looking for 1.4's code specifically? (other than...1.4 was incredibly useful compared to the newer ones IMO)

[ShadoW_86]

1.4 is best GtkR ever, reason enough for looking for it . But actually, I didn't check 1.6 in long time, is there anything really interesting added/done to it?

[Noruen]

Why 1.4? What's wrong with my life love GtkR 1.5?

[Eraser]

I've never been able to get 1.5 to work properly so I've always stuck to 1.4.
Never bothered looking for 1.6 though.

[Delirium]

because 1.5 and versions onward have such shitty brush selection in both the grid and camera view, I really love how 1.4 handles brush selection :<

[VolumetricSteve]

I maintain that someday I'd like to build a java-based radiant that would take the best features of gtk 1.4 and netradiant, and then add in some new features of my own, I just need to get my head around a lot of the java GUI stuff, but yeah, 1.4 was good, I don't know why they messed with the interface so much in the later versions.

[obsidian]

java-based radiant

Please don't. Java = (IMO)

If anyone wants to contribute anything code wise, please help out with 1.6. It's the only one official GtkRadiant that's seeing active development and it needs more talented programmers. It's built off of the 1.4 branch and if things go well, certain 1.5 features can be ported over with a bit of elbow grease. There's been some positive activity recently, join the mailinglist or I can hook you up with the right info. I'm running the latest binaries from yesterday to help debug (website binaries are old).

[VolumetricSteve]

No java? I was just thinking that, particularly for mac users, getting any radiant to work is a pain, NetRadiant only works easily in Leopard, and if you so much as look at snow leopard funny, net radiant breaks. It crossed my mind "this stupid piece of !@#$% wouldn't constantly have OS problems if it were already cross-platform"

What do you have against Java?

[obsidian]

It's not C that is causing the problems for OS X support, it's just that it hasn't been updated for OS X in so long and there are no active OS X programmers and testers working on Radiant. There are probably some GTK+ toolkit issues for OS X as well and Apple has been doing funny things lately with regards to legacy support. But simply converting to Java won't fix anything and Java may run slower off of its VM.

If you are an OS X developer, you are better off contributing to the current code base and just fixing the OS X bugs than doing a long winded conversion of the whole thing to Java. It's like you're trying to fix a leaking faucet by piping in an entirely new water main direct from Alaskan melted snow.

[VolumetricSteve]

Yeah, I just felt like that's what it needed, at least in OS X, the problem always comes down to X11, and the problems never seem to really go away, some things could just be fixed with bug fixes, you're right about that, just X11 is really old, and poorly implemented. I thought the OS X version needs to be taken away from X11 and ported directly into Cocoa, then I thought about how complicated that would be, and at that point I figured "why not just find an old java version, update it, and be done with X11 forever, and go cross-platform" but these days, I'm more and more frustrated with OS X and later this week I'm planning on migrating to linux so...I don't know, maybe the linux version of NetRadiant/GTKRadiant will be amazing. None of this is here nor there, in answer to the original question, a good long while ago I was downloading and archiving as much quake 3 related stuff as I could find on the web, and nowhere in there was the source for 1.4, unfortunately.

[Eraser]

java-based radiant

Please don't. Java = (IMO)

Typical comment for the ignorant.

[Noruen]

Well, I also heard that Java is slow (OpenOffice ala LibreOffice is good example), but I'm not a programmer

And I've got a great idea - isn't it possible to include support for UnrealEngine 3 into Radiant? Or some plugin for UnrealEd? Because method of level-building in UEd is so stupid I can't describe (you know what I mean).

[Eraser]

I think the whole methodology of level building in the Unreal enigne is opposite of that of Quake 3.

I have no experience with this though, but I understood that in Unreal, you basically carve your map out of one huge brush so to speak. So instead of adding brushes to create a closed off map, in Unreal, you extract sections out of the void which become passable areas.

Not sure if that would be possible in GtkRadiant (well, anything's possible but not sure if it's worth it)

[Eraser]

Well, I also heard that Java is slow

Depends on what you're using it for.
I think what Obsidian means is that Java is crap for real time applications such as 3D renderers. While that is partially true (C or C++ is a much better option in most of such cases), a general comment that "java sucks" or something like that is just plain wrong.

[Noruen]

Yes, but I think there are two possibilities in building levels - substracted or "additive" (like in GTK). Bu definitely it is not what's making me angry - it is that method of "one working brush" you must move in map. Simply say - what I can create in Gtk in 5 minutes I can do in UEd in 5 hours (just overshoot, but it correspond with my feelings).

About Java - and isn't it because of fact, that C++ is compiled into "machine code" and java is zip archive with files?

[Eraser]

About Java - and isn't it because of fact, that C++ is compiled into "machine code" and java is zip archive with files?

Uh, yeah sort of.
C/C++ is compiled to native machine code and Java is compiled to bytecode, which is then interpreted by a VM, or "virtual machine" (a piece of software that runs on your computer, called the JRE) into machine code. The upside of this is that your Java program can be ran on any system that has a compatible VM running on it. The downside is that this interpretation step takes time and system resources and is thus inherently slower that code that's compiled to native instructions.

[spirit_]

The speed difference between Java and C shouldn't matter much for an application like Radiant but obsidian is definitely true with his other argument: it's insane to try to re-implement everything in Java instead of helping to improve the existing tool. It's even open source!

@Noruen: UT3 has a very different approach to level building and it would most likely take ages to make something that is more suited to it than the UT3 devkit. Since the Radiant guys seem to have trouble keeping the editor alive I highly doubt that they have time for that.

[Noruen]

Oujéé I have had better knowledge than I thought Thanks.

spirit_: yes, you're true. It is just my dream. I just wanted to say that Gtk Radiant is the most user friendly editor i know. I'm very curious what "Rage" will bring. I hope RadiantStyle will be maintained. If not... goodbye my mapping part of life

[VolumetricSteve]

Whooooaaaaaaa, didn't think all this would happen. I was just thinking that since Radiant is really just a GUI, not a performance application by any means, that Java could be used to avoid all of the various Radiant's platform related issues. Plus, GTK is a huge library and it's pretty lame in Windows, especially when MFC was available at the time Radiant was being worked on originally. I saw that Radiant is a big 2D interface paired with a generally smaller, but not limited 3D interface, and a window for selecting textures, and some dialogue boxes for additional tasks. All of this can be done in java now, fairly easily as far as I know. It'd remove the reliance on GTK in windows and linux, it'd remove the need to ever think about X11 in OS X again, and only one version of the code would have to be updated, not one for each major OS. Anndddd whenever Silicone Milk's Q3map3 becomes available, I was going to try to see if CUDA could be layered in with Java, so you could see more of what your level would look like in the editor with lightmaps (in real time?) similar to how the Doom 3 level editor showed you volumetric shadows in real time. I realize this is overly ambitious and crazy, but...that's kinda what I do (unless there's some insurmountable issue in the way).

This thread really went off the rails, sorry Delirium.

[Eraser]

Developing something like Radiant in Java would be no problem. However, the real question is if it's worth it. What you're looking at is developing a complex application from scratch which means you'll be running into lots of bugs and problems that are inherent to developing new software which were all ironed out of GtkRadiant a long time ago.

I don't want to discourage you, but I think that recreating something that can match GtkRadiant 1.4 will take a long, long time. I know you don't want to hear this, but obsidian is right in saying that all this time would be better invested in improving Radiant 1.6

[Delirium]

lol, anyyyyyyway.

So its a high chance that noone would have a 1.4 source anywhere? I don't recall a official release by ID.

reckon I'd have any chance posting on other forums? (Splashdamge?)

[VolumetricSteve]

@Eraser: No, you're both right, and I agree, it would take a long time,but part of my ambition is also in building something from the ground up so I can document it as well as I can so people can easily build on it for different games/applications. I don't have the energy to read through all of GTK 1.6's code and try to understand all of the design choices people made in coding it it over the....last decade and a half Radiant's been in progress. I do have the energy to throughly learn Java (something I've been wanting to do anyway) and use that knowledge in a positive way that'd be beneficial to a bunch of people, and create a learning tool for other people in the future.

@Delirium: I'm about 95% certain you won't find the GTKRadiant 1.4 Source anywhere, as I had a wild and wacky weekend a while ago where I downloaded a TON of quake 3 stuff and built my own little Quake 3 Archive incase the world ends, and nowhere in there is the source for 1.4. I'll take another look when I get home though.

However, NetRadiant is basicially a more reliable version of 1.4(I think), even though they report their latest version number as 1.5.0. Perhaps you'd wanna start there? Good luck.

I'm pretty sure this isn't it but it's as close as google is getting me:

https://zerowing.idsoftware.com/svn/rad ... nches/1.4/

[stevec]

Netradiant = based on Radiant 1.5

Personally I found all versions of Radiant terrible when I was a bit younger, the installation process made a complete mess and I was but a small-minded Worldcraft/Hammer kiddy.
My first encounter was with... Ion-Radiant...
It is the tool that the levels from Daikatana were forged in. This helped create an anti-radiant bias.
I actually ended up writing some tools that ported content from Anachronox (shared the same editor) over to Half-Life friendly formats so I could avoid using Ion-Radiant.

Then I grew up and found NetRadiant.
A portable version of Radiant 1.5 with many, many bugs fixed and an improved q3map2.
After adjusting over a couple of days I was finding myself enjoying editing levels for the first time in years, it was also 'flowing' better then Hammer ever did.


Oh, I'm not sure about the 1.4 source... Did ID ever release it?
I think 1.6 is a continuation of the 1.4 source with features back-ported from 1.5.