Leaderboard

Happy Halloween! (if that's an important thing where you live, actually not where I live ) This is from my newest project, unfortunately extremely unlikely to be released before Christmas 2023, maybe even later.

I finished adding all the "works" that are free for adaption.

Getting close to what I want, just need to finish implementing notes and rating, it will read the FMS folder and gather the info and will launch whatever FM is selected fine.

@AluminumHaste@SeriousToni This should be fixed in source now. The automated build is running on Github, you can grab a portable package when the build is done (scroll down to the Artifacts section): https://github.com/codereader/DarkRadiant/actions/runs/3358810112

I played a T2 mission with a timed feature in it You had to throw a switch to open a bank vault & then get in the vault before it closed, I've hidden the details just in case anyone hasn't played it I eventually did it but it was more blind luck than skill & I was seriously hacked off with the mission by then Put timed events in missions by all means it does add to the challenge but that T2 one was ridiculous --EDIT-- I had forgotten what this one was called & accidentally loaded it up again & got all the way to the vault door before I realised it was that one I deinstalled it, I'm not doing that stupid bloody jump again

Btw. Wreckers Reach does not load in 2.10+. But I got it working by removing entity 4045 and 4046 from wreckers.map using a text editor. I added this info to the wiki page. Nice map btw.

Yeah, I always do an extended search for tools from The Netherlands first. Everything local is always better. If not available I am forced to use garbage from other countries..

I use ShareX, also FOSS and the most complete, but sadly only for Windows. Also a very fine app is Flameshot, cross patform and also FOSS

I just found this awesome cross-platform software to make and edit screenshots: https://github.com/ksnip/ksnip I think it's very useful for people beta testing missions and other kind of feedback based on screenshots. You can load in existing images (for example TDM screenshots) and edit them. Edit: It also has upload capabilities: Would be nice if you could also upload to Google drive folder for example (or Dropbox), or Postimage via custom script.

I don't have an issue with reusing assets (like scripts et al.) either, reusing whole map parts would be odd, though. But if someone says, whoa, that's a nice doorframe or alters for example the puzzle in the Old Habits Rebuild, I see no issue. It never hurts to ask, though. In regards to absent authors it is probably better to not reuse their stuff without consent. It just creates drama in the worst case and in the best case mission authors save a bit of work which they in return have to invest to understand whatever the previous mission author has bashed together. EDIT: @Dragoferpretty much nailed it.

My view would be: If the author is still reachable, ask for permission unless they already gave blanket permission. If the author is not reachable, check whether they stated anywhere that they're opposed to their work being reused (i.e. in the FM credits). If not reachable and no statement was made, I'd say the work can be reused if you give abundant mention of the author in the credits in the release thread and in the FM package. The vast majority of authors (who don't explicitly say they forbid reuse) are probably fine with that. I'd distinguish between reuse of assets and map sections, however. The latter can easily be perceived as plagiarism, so it should always require close coordination with the original author if you want to use parts of their map in your own.

1. If you want to use custom work (sound, models, textures) from released missions, best thing is to ask the author if he/she agrees. So far, Sotha is the only one who permits unlimited usage of his custom work (he expressed that here on the forums). He just wants to be credited. 2. If mission authors from released maps are not available anymore, then I do not know what to do. @Dragofer@nbohr1more: Any suggestions? I would say, that's the same situation as described above, just replace "released mission" with "abandoned work". As for my released work, feel free to use whatever you want.

Yes, Ive just sent the link to the current campaign archive to you 2.

When I add more headings it adds that automatically. I just did and it shows up now.

Ok, but do you have the whole "work" because Fllood doesn't have it.

I think it was a small part that I split off from the main map because NeonsStyle was only interested in a specific aspect. It should be ignored. Brush/patch/entity count is an interesting statistic to have, too.

A table of contents at the top would be helpful.

I saw there was a download somewhere in this thread called neons_harbour.zip . Is that the correct one for Wrecker's Reach and why was it renamed?

A first draft of the wiki page: https://wiki.thedarkmod.com/index.php?title=Abandoned_Works DarkRadiant screenshot(s) need more info?

Collision gets generated from the perspective of the AI or player that will navigate it. If you create complex brushwork with tons of triangles, then all that geometry must be evaluated to determine how it might impact physics and navigation. Doom 3 was designed to be mostly indoor corridors so most brushes would be low-poly flat walls. Places in Doom 3 where geometry is more complex tend to be made with Modelling apps like 3DS Max or Lightwave though some complex geometry is in the form of patches or brushes converted to func_static. In those special circumstances, clip brushes like monsterclip, playerclip, or caulk are used to ensure that no 2003 era CPU need to perform 3D physics evaluation of these high-ploy structures. The same holds true in TDM though in 2.11 we do have some new model BVH optimizations that make it less critical to use clip brushes. As I recall, these optimizations only apply to the real-time performance of a mission. Dmap still needs to generate collision surfaces and that is still very difficult. TDM's dmap is far more optimized than Doom 3's but it's still vulnerable to complex structures with lots of connected edges needing to evaluate a huge number of permutations to resolve the final optimized collision structure. If you have a powerful enough CPU, lots of RAM, and possibly a decent chunk of patience for compile times you might be able to dmap a pathologically high-poly mission just by waiting 30 minutes or more. It's not an advised practice though. Map loading still has to parse the compiled collision data and even the optimized result might take too long or eat too much RAM during the loading process for many players ( see how many folks couldn't load "The Painter's Wife" or "Penny Dreadful 3" when they came out. ). We still recommend the use of clip brushes to simplify collision and path-finding just as was done in 2003. A few relevant topics: https://wiki.thedarkmod.com/index.php?title=Pathfinding https://wiki.thedarkmod.com/index.php?title=Performance:_Essential_Must-Knows https://wiki.thedarkmod.com/index.php?title=Systematic_Method_for_Adding_Pathfinding_to_Uneven_Terrain

If you set models to non-solid, they won't affect collision detection. This is a good idea for models used as decorative extension, like little beams on walls or other small stuff (like static junk). It won't speed up aas creation, but the cm file is created faster and it becomes less likely that ai gets stuck on small objects. Obviously you should't do this with large objects.

By this you mean the back, what is not seen by the player? Otherwise I don't understand, probably have to do some reading on the subject. Edit: Sorry this goes a bit off-topic. If it gets longer, maybe it needs to be split into a new thread.

As long as you are not working on optimizing gameplay or ai, you can use noflood as argument for dmap (dmap noflood mapname), this will skip the aas creation (pathfinding). As @nbohr1moresaid, the pathfinding algorithm or more precisely the time it requires isn't linear with the amount of areas the ai can step into. If the map isn't properly monsterclipped and a lot of decorative stuff made from brushes is left worldspawn, you can easely end up with a huge workload. The same applies for the collision data creation.

Probably needs monsterclip. Dmap is probably choking on determining complex collision for AI paths. Block out caulk for the player to walk / mantle on and block out monsterclip for AI then set most of the map geometry to func_static and non-solid

Just want to report that this map loads fine in tdm 2.11 dev. Dmap takes ages though, so I quit for now. During dmap it shows: Merge leaf nodes and this process becomes slower over time. Anyone knows what merge leaf nodes means? Fixed 8 out of 10 visportals in the hope that it would help somehow, but it did not. Edit: When I remove all Ai it dmaps quickly. Pretty nice looking map actually.

No, I returned it a while ago after making Perilous Refuge from part of it. NeonsStyles last expressed an interest in it.