Search the Community

Hi all! I wanted to share some thoughts on DarkRadiant tweaks & features. I’m guessing most of this has already been mentioned before, but I wanted to throw it out here just in case! Thanks. Particle Editor Currently, it’s a bit difficult to judge a particle effect’s size relative to other particle effects, as the camera automatically repositions itself to frame the effect in a close shot. (The ground grid does give a frame of reference, but is only useful for very rough estimates.) I’d love it if there was a way to “lock camera position” while viewing particles. This would allow me to view a series of effects and more easily identify which effects are small, large, etc., as the camera position would be locked in place. Texture Wrapping It would be greatly useful if there was functionality to “wrap” a texture around a brush, as described in this video: https://www.youtube.com/watch?v=aBvj-d2AiT0 Ideally, this would help automate a lot of the manual processes involved in texture angling and alignment. Texture Browser For the “Choose Shader” window (S > folder button), I’d love it if there was an option to include thumbnail visual images of each texture next to each texture name. Having it be image-oriented would make it easier to browse textures at a glance. Also, it would be helpful if the texture dimensions were displayed, either on the thumbnail or in the description. Vertex Editor Currently, the editor prevents brushes from becoming concave during vertex editing. It would be helpful if there were an option to temporarily allow concave brushes during vertex editing, as the current convex-only safeguard sometimes makes it difficult to do more complex vertex edits. Also, currently if I drag a vertex on top of another vertex, the the entire brush is immediately destroyed. It would be great if the vertices were instead merged, and automatically collapsed the missing edges/faces. Three-Point Clipping When using the clipping tool to cut along three points, it appears that when I move one of the three points in a given orthographic view, its coordinate in the inactive axis will be reset to zero (i.e. if I move a clipping point in the X-Z view, it will reset to Y=0). This current setup makes it difficult to do certain clipper cuts -- it’d be a big help if the clipper didn’t reset to zero. Entity Inspector Currently, when I have multiple entities selected and open the Entity Inspector window (N), the entity key fields are all empty. It’s a little difficult to view how entities differ, or to apply batch changes (i.e. the workflow is select one entity > decide which single key-value pair you want to change > change it for one entity > select all entities you want the change applied to > reapply the change). It’d be great selecting multiple entities showed a union of all the properties belonging to the entities -- with common values showing the key/value, and with conflicting values showing the pseudo-value of “different”. ReloadCurrentMap When using the in-game editors (i.e. ragdoll editor, light editor), the game saves over the current .map file with the changes made. It would be great it if there was a DarkRadiant keybind to reload the currently-loaded map, to more easily load in the changes made by the in-game editor. Right now I’m manually reloading the map via DarkRadiant’s File menu, but a keystroke shortcut would help streamline this. Editor_usage text It would be nice if the editor_usage help text appeared somewhere in the Entity Inspector (N) window when an entity was selected. Arbitrary Reflection It’d be great if reflection/mirroring was allowed around an arbitrary plane (i.e. if we need a symmetrical detail on a wall that runs at a 45 degree angle). This is currently possible with the existing tools, but it’s a bit tricky, and in the case of more exotic angles, oftentimes forces us to go off-grid. Scripting The scripting interface is very powerful and flexible, and has been hugely valuable. Something that would be great is some way to operate on selected components, i.e. a way to operate on specific faces, edges, or vertices. Currently with SelectionVisitor, we can traverse a component selection with GlobalSelectionSystem.foreachSelectedComponent(). However, the nodes traversed with SelectionVisitor seem to be the brush the component belongs to. Ideally, it’d be great to be able to traverse selected components. Also, there seems to currently not be a way to identify components (i.e. there’s an isBrush()/getBrush() method pair, but not an isFace()/getFace()) -- admittedly, not sure if I’m fully understanding how this system works, so apologies if I’m misunderstanding it!

Probably the .script file that's needed, but unfortunately can't be included with a prefab. Maybe there's a comment somewhere in the prefab (or in the wiki or forums) that says what script it needs and where to find it. (The similar combination setup I used earlier doesn't involve "slot", so clearly uses a different script that I what I have.)

Since it already has been mentioned in the forums, I'd like to announce a small project of mine. A few weeks ago I was playing Legend of Grimrock and it struck me that their level design is extremely simple and modular, and yet, the player can spend quite a lot of time in the game. And it looks pretty, too! So I wondered if it wouldn't be possible to create a modern "Dungeon Crawler" (solve puzzles, fight monsters, collect loot, progress your character, upgrade your equipment) type of game inside the TDM engine. The engine already supports a lot of things one needs for this, and the overal structure and assets work, too. And with prefabs, one might get the level layout done quickly. However, building a few test prefabs in DR is easy, but creating a full mission out of them is quite painful. Not only needs it a lot of planning, but you can also spend a lot of time "upgrading" things later on. For instance if you later want to add a grime decal to the walls, you have to revisit the entire map. Even worse is if you find out later that your block size must be bigger or smaller. So the idea was born to create a sort of framework that can assemble missions from prefabs. Preferable while getting the description of the mission from a text file. So far, this has been a lot easier than I thought. Here is what I got working so far: Overview: You can describe your mission in a (Unicode) text file. This contains overall options, different locations (each location can have its own ambient light, music,name, fog), and the connections between the locations. Each location can have multiple "floor levels", these are stacked on top of each other. The config file also specifies which symbol means "use this prefab". It is also possible to specify links (per location), which means you can say "this lever with the symbol A opens the door with the symbol D". The framework reads the prefabs, and then positiones them in the map. It also glues all the locations together, adds location_info entities, a player start, an exit, and an objective to reach the exit. The resulting map is then enhanced with script objects (all nec. assets are bundled together), and automatically dmapped via TDM. Everything then is packaged together into a working .PK4 file. My demo map takes about 20 seconds, where 15 are dmap. In addition to the "basic" stuff I also managed to get a few things working, like a pressure plate, portcullies, and also made some puzzles. Oh, and per location fog (fading from location to location). Different difficulty levels are also supported, one can specify "this prefab appears only on easy" etc. You can find more info and screenshots and demo here: http://bloodgate.com/swift/ There is also a developer diary where I will be posting interesting entries from time to time. Here is an DR shot of a sample level, consisting of small modular prefabs and one large (the large hall on the lower left): The next steps will be to add more randomness (either static at map generation, or at runtime, so the map is slightly different each time you replay it). Also, while it is already possible to "overlay" prefabs (e.g. "for this location, look first here before falling back to the default"), it is not yet possible to "reskin" prefabs. This would be something which is impossible in DR (you cannot really reskin worldspawn brushes, unless you live with the fact that it is all manual For now I'm quite excited!

I'm trying to fix the warnings on my end, but remodelling the models is out of my league, so you're going to have to fix the stairs model on your end: "WARNING:ConvertLWOToModelSurfaces: model 'models/darkmod/architecture/stairs/set01_stairs.lwo' has 17/8108 nontriangular polygons. Make sure you triplet it down" Edit Also there's this model, which is probably a modelling issue as well: "WARNING:ConvertLWOToModelSurfaces: model 'models/darkmod/misc/clipmodels/pickaxe_cm.lwo' has bad or missing uv data" I still have hopes for script-hotfixing the following warnings: "WARNING:Couldn't load image: tdm_tongue [map entity: atdm_ai_townsfolk_female_1] [decl: atdm:ai_head_female02_base_brunette in def/tdm_ai_heads_springheel.def] [decl: female_head02 in def/tdm_ai_heads_springheel.def] [model: models/md5/chars/heads/npcs/female_head02.md5mesh] [decl: tdm_tongue in <implicit file>] [image: tdm_tongue]” (Edit: I hotfixed this issue and attached the hotfix in the tech support forums. The "tdm_tongue" shader should simply be renamed to "tdm_character_tongue".) “WARNING:Couldn't load image: models/darkmod/wood/boards/wood_brown_dull01 [map entity: func_static_53] [decl: old_plaster in skins/tdm_models_architecture_modules.skin] [decl: models/darkmod/wood/boards/wood_brown_dull01 in <implicit file>] [image: models/darkmod/wood/boards/wood_brown_dull01]" (Edit: I hotfixed this issue as well - see the tech support forums - but I think you should still go over this file, because there are a lot of other suspect textures in there as well.)

Thank you for a great puzzle fm! I'm having real trouble with the TDM v 2.10-64bit on Linux Mint 21.1 Link to the guide supplied by V-Man339 https://forums.thedarkmod.com/index.php?/topic/15844-fan-mission-the-gatehouse-by-bikerdude-goldchocobo-updated-02112014/page/7/#comment-429405

I think that the existence of a dedicated bug tracker for this engine proves that I should somehow make it known when it crashes. I just figured that the tracker was an internal thing between developers, and that it was okay to instead report the bugs here, especially since I desperately needed help with it, and since you have two sub-forums mentioning that they're intended for reporting bugs. ...so why, when I report a bug, am I getting shamed for it? Why was the guy in the thread you linked getting shamed for not backing up his project, instead of the main dev acknowledging that there is a very deceptive button in the engine, that can remove fan mission folders when you press it? Blaming the reporter for not backing up his work, is certainly not how I would personally handle reports of entire projects disappearing, but I guess now I'm warned about how he sees these things, to the point where I'm wondering if it's a good idea to report any bugs I find in the future, at all, if I'm just going to be considered a bad person for it. "Hm, to my experience, crashes in free software is a regular thing. Try "Natron" and you will understand what I mean." I don't understand your point. If crashes happen frequently enough, should we treat them like Covid and just "try to live with it"? I'm pretty sure that the bind bug is easy to fix, but it's like you're saying that bugs have a right to exist too. "Lots of longtime DR mappers ask questions in the Newbie thread (myself included). Nothing to be ashamed of and by DR/DM standards you are a newbie (and you have not to be ashamed of that as well! )" So if I've programmed for 30 years, mapped for Dromed for countless years (since roughly the turn of the millenia), mapped for the Dark Mod for 1 year, but joined these forums just a few weeks ago, I'm a newbie. I think that by that definition, the word "newbie" has lost all meaning. Besides, reporting crashes are not the same as just asking questions. Anyone can reply to a question. It takes a developer to fix a crash issue. ...and it's like you don't expect the developers to even fix crashes, and I really hope that you're wrong about that. For example, since I updated to this year's version, Dark Radiant itself, hasn't even crashed once, and it used to crash all the time, and so I take that as evidence that development involves fixing crashes. "If you want to report a bug, then please use the Bug Tracker." Will do. I was just fooled by the subforum descriptions actually telling me to report bugs in them. "You wouldn't believe how many mappers lost WIPs because of defective hardware" Again: I don't see your point here, where you argue for not fixing defective hardware. When I'm told about a bug in my program, I just go "Thank you. I'll fix that.". I don't go "Well, weren't you a sucker for using my program.". That's not how I was brought up at all. "for somebody who expects Armageddon in the near future (and there are really good reasons for that), you have a pronounced need for communication. I think that does not fit, because: Why go on with communication when all hell will break lose the day after tomorrow?" Well, why not? Yes, I am so interested in mapping, that it's actually cutting into my apocalypse survival routines a bit, but that's just human nature, and a month's vacation from prepping, won't make that much of a difference. "I hope that you - despite of your own assessment - will finish your WIP and we all can enjoy a cool new mission in the future with your name in the credits." Well, for me, mapping is about pleasing myself - not others. Play The Beginner's Guide, and you'll see what I mean.

I agree. I'm copying this idea over to the https://forums.thedarkmod.com/index.php?/topic/21741-subtitles-possibilities-beyond-211/ thread. Related: I'll probably do some more experiments with the current TDM font & size, to understand max char count versus field width.

Hey, sorry for the long wait, I tried to delete the Darmod.cfg completely and let the game create a new one upon starting, and the issue persisted. I captured a few screenshots I will attach under the spoiler. they are all captured inside the arsenal building, two up at the kitchen, the other two in the basement, near the grate with the arrow next to it. It might not be obvious on all pictures, but I've tried to capture the reflection of the moon in the skybox inside a building that had a very clear defined roof above the surface with the reflection.

I added material for the ground. Here's the new link. I'm pretty sure there is a light? As for nodraw skin, I use it because player's reflection is ugly. It doesn't affect arms and weapons. There's one element of the player model still visible in the reflection, probably the hood. I'm sure there is an easier way to make the player reflection invisible? Couple of things you have to consider when using real time reflections: they are buggy sometimes and they are computationally expensive. Also you need two surfaces with two different shaders: one for reflection and one for refraction right above the first. The second one uses a large set of tga textures which may also affect the performance. For the reflection, one can use cubemaps instead of real reflections but they don't look that good without parallax correction.

Some years ago i succesfully downloaded the wiki with wget, for TDM dvd. I dont know if this method wikl work now. https://forums.thedarkmod.com/index.php?/topic/19998-tdm-collection-dvd/#comment-437887

I guess the best image-to-normal conversions I've seen here in the forums are via njob. I am curious about this AI thing though: https://github.com/HugoTini/DeepBump has to be installed into Blender as a plugin?

I'm seeing some weirdness in the 2.08 final regarding reflections. I've seen it in two different missions that I'm beta testing. Here is an example in HHTLC, in 2.07 it looks like this, a nice mild reflection (the plant pot): But in 2.08 the reflection is wonky: Same graphic settings. I deleted the darkmod.cfg, just to see if that made a difference - nope. I saw this problem in another FM too. Is anyone else seeing this?

No, not really, have only seen him on the forums many years ago. He deserves a lot of credit for providing the SVN infrastructure in the beginnings of the project, it's been only later when we transferred this to a hosted server. Memory is blurry, but I think my time working on TDM and his don't overlap that much - I had been getting more active, and he pulled back a bit. He bootstrapped quite a few important systems, IIRC he's been working on light gem and the first Stim/Response and Inventory implementations. From what I recall, he gave the project an organisational backbone in the technical department which is crucial to keep things together. Folks like Spring and NH who have joined long before me could give more insights, I guess. (Looking at my join date makes me feel old either way. From TDM's current view point, the year 2006 seems like right at the beginning, but actually the mod had already been existing for two years by that time I joined. With the first release in 2009, 2006 is rather in the middle. Heck, I haven't touched the mod code for at least 10 years - and I can still remember a few things, which is a testament of how much it occupied my thoughts back then).

There are other complications though. How much fall damage should player take, if they decide to jump off a rope with the body? Should the player let go of the body or not? Also, right now it's much harder to jump off the rope with the body than without it. Why? And last but not least, how would you teach players these things, possibly without much hand-holding and text prompts explaining the rules? I guess I'm with @STiFUon this one, if you restrict dropping the body, you'll save yourself (and mappers) a lot of headaches. But even that doesn't solve all the problems, I know I'm in the minority in these forums, but as a player, I really appreciate the beauty and efficiency in simplicity of the design. Not overthinking everything and adding more and more rules for the sake of realism (or anything else).

I don't think it is possible to have a dynamically-rendered reflection that doesn't increase shadow rendering, because the reflection is effectively a completely separate render from a different perspective (i.e. the other side of the plane looking back towards the camera). Assuming you want correctly-rendered shadows in the reflection, it is going to be necessary to render the shadows from this perspective, and there is no way to re-use the shadows from the main render because it's a completely different camera setup with a different field of view. The only options, as far as I know, are the fully-dynamic reflection which has a performance hit due to the second render, or use a pre-rendered cubemap which is much faster but does not respond dynamically to changes in the scene.

This is basically "do include my work ASAP because I worked so hard, or else *sulk*". This is similar case: https://forums.thedarkmod.com/index.php?/topic/21679-beta-testing-211/page/10/#comment-482352 This is neither a commercial product, nor a phishing email. That sense of rush and pressure is artificial. These releases typically do take long, and even then, there are often many things broken by mistake or omission. Often there aren't enough people to test stuff, or they're not competent enough, etc, etc. There's little point in hurry.