I made some changes to the DarkRadiant Video Tutorials wiki page. I added missing descriptions to some of the video's and more video sources.

https://wiki.thedarkmod.com/index.php?title=DarkRadiant_Video_Tutorials

Just wondering if people might disagree with some of the changes, especially the added video's from different makers. Or people have suggestions. If not, then good.

Edit: I removed the link to Bikerdudes video's again because in them he talks of reusing parts of maps from other missions (in this case I think mission A new Job). I didn't check out all of his video's in that playlist, so I didn't spot it before. The reason I removed them is that I don't want people to show the wrong example of building. Also, I think that the video's (although very informative), aren't great purely as instruction.

Edit 2: Added dates of video submissions.

and the general discussion after it came to mind) about the usability and efficacy of Flashbombs in TDM.

To wit, most people seem to think they're kinda bad because they're just a stun, which still requires you to hide in a very narrow time frame and then wait for the AI to cool off after use. One of the big upsides of flashbombs in TG/T2 was that, if you had alerted human enemies chasing you, you could drop a flashbomb and then turn around and blackjack them to non-lethally remove them from play. It was an inventory-limited opportunity to recover from failure and continue playing the game, as opposed to just reloading a save. This is a fun and proactive interaction, and I propose that we add it to TDM.

Specifically, I think this can be accomplished with minimal code. After a read through the public TDM git repository, I think the most appropriate change would be to adjust the condition here: https://github.com/stgatilov/darkmod_src/blob/ac0a286561630eefee1cbb44d09d77128cd3d8e7/game/ai/AI.cpp#L11892 to read as follows:

  if ((GetMoveType() == MOVETYPE_SLEEP || // grayman #3951
         GetMind()->GetState()->GetStr() == "Blinded") && // proposed - maybe there's a better way to write this condition like checking the type or something?
        ((minDotVert != 1.0f) && (minDotHoriz != 1.0f))) // cos(DEG2RAD(0.0f)) indicates elite faceguard helmet
    {

Currently, this check does not pass for blinded AIs and they move to the if-else branch at L11903, and since they're very alert because they were actively chasing the player, they cannot be blackjacked. This change should give a flashbomb-blinded AI the same knockout vulnerability angle as a sleeping AI. Helmeted human AIs (and undead/magical AIs because they can't enter that particular mind state) retain their blackjack immunity, and everyone else can be clunked in the face as a reward to the player for spending a limited resource, not flashing themselves, and having the quick thinking to turn around and draw the blackjack. The blinded mind state lasts for about 8 seconds (I forget where I found that but its a hardcoded magic number in a Damage() function somewhere), which feels a bit short but about right, and then when the state changes the player's window of opportunity is lost.

My brain kinda glazed over when I looked at the SVN checkout+compilation.txt instructions, but if I can help test or debug this with a little hand holding I'd be happy to do so.

Please I need help!!

Thank you.

it likely isn't vertex normal related since all of the outside is blue

and the model looks fine in dark radiant, I can provide more information about the models and things if it is needed

In my spare time I like to walk around, record sounds (field recording), make dark ambient and experimental music, and code. My dark ambient music is available for free at https://wolfmond.bandcamp.com

I'm currently working through the Youtube Tutorial from @Springheel and I'm really enjoying learning level design.

I look forward to getting in touch with you and contributing my skills to the community.

Best regards, Wolfmond

___________________________________

Hallo  

Ich heiße Wolfmond, bin 41 Jahre alt und komme aus Norddeutschland. Ich bin euer Neuzugang. In meiner Jugend habe ich mit großer Begeisterung Thief: The Dark Project gespielt. Vor kurzem habe ich The Dark Mod entdeckt und begonnen zu spielen und bin beigestert, dass es immer noch eine aktive Community um dieses schöne Spiel gibt. Mich interessiert vor allem das Leveldesign und die Erstellung eigener Kampagnen. 

In meiner Freizeit gehe ich gerne durch die Gegend, nehme Geräusche auf (Field Recording), mache Dark Ambient und und Experimentelle Musik, und programmiere gerne. Meine Dark Ambient Musik gibt es kostenlos auf https://wolfmond.bandcamp.com

Aktuell arbeite ich das Tutorial von @Springheel durch und habe große Freude daran, Leveldesign zu lernen.

Ich freue mich darauf, mit euch in Kontakt zu sein und meine Fähigkeiten in die Community mit einfließen zu lassen.

Viele Grüße,

Wolfmond
 

UPD3: Fixed Russian fonts can be found/downloaded here

- and, yes, - we're working on subtitles and on translation of all audio files;

UPD4: 62 missions translated and counting; Public change log;

Sill missing testers, but the first reviews were quite good.

At this point i'm kinda ready to admit that...

UPD5: at this moment we have 76 translated missions

UPD6: 90 missions translated and counting...

UPD7: 99 missions translated, tested and uploaded on darkfate.org

The ebuild itself is here: https://git.sr.ht/~dblsaiko/ebuilds/tree/master/games-rpg/thedarkmod, and you can install the game by enabling my Portage overlay and then emerging the games-rpg/thedarkmod package:

# eselect repository enable 2xsaiko # (alternatively, add the repo using layman)
# emerge games-rpg/thedarkmod

This fix will center the UI for any resolution wider than 16:9. For resolution that are more square i.e. 4:3 and 5:4 the UI will be anchored to the bottom instead as having the bottom elements floating in the middle looks strange.

I also had to employ a trick to make some fullscreen effects stay fullscreen instead of becoming letterboxed, such as the transparent black backgrounds when reading books. Any rect that is 0,0,640,480 or larger will be considered fullscreen and stretch as normal. Of course some are note supposed to stretch like the custom main menu backgrounds so I added a bool variable for windowDefs called aspectAware that can be set to override this behavior. I experimented doing the opposite but could never find out how to make the book backgrounds stretch so I went with this.

Still haven't figured out how to position the compass correctly but it looks decent enough.

I've been working on porting The Dark Mod natively to FreeBSD. I'm happy to report that at this stage, I have a working thedarkmod.x64 and tdm_installer executables on FreeBSD 13 amd64 (x86-64) with SVN revision 9889 and 2.10 game files. I've been able to play through both default missions as well as several contributed missions.

There were minimal changes required to get a working port, including:

1. Fixing the build with Clang, FreeBSD's system compiler (committed in 9889 by Stephan)
2. Fixing up header files and #ifdef's for FreeBSD (attached here)
3. Fixing up header files for zipsync (attached here)
4. Modifying the CMakeLists.txt to use FreeBSD system libraries instead of the pre-built libraries for Linux (working, but still cleaning this up before upstreaming the changes)

I hope the portability changes will be accepted upstream. I'll also look to see which CMakeLists.txt changes would be appropriate to upstream, but it is not too much of a problem for me to maintain it locally for FreeBSD.

Once most of the changes have landed, I also plan to create a package ("port") with a thedarkmod and tdm_installer binary.

Regards,

Kevin

patch-FreeBSD2.diff patch-zipsync.diff

I understand that TDM assets use the CC-BY-NC-SA license.

This trivally means that for creating these assets we can use anything licensed under the same license, CC-BY-NC-SA, or anything under CC0.

What about other CC licenses? From this page in the CC FAQ I understand that we can also use CC-BY and CC-BY-NC, as those can be relicensed under CC-BY-NC-SA. So far so good.

I have noticed, though, that most asset libraries today (free or paid) use some form of "roalty-free license" or "asset license" which allow any type of usage without attribution, even commercial, except redistributing them in their original form. This includes some of the asset libraries that have been link around this forum as usable. This licensing is most prevalent with models, but for example textures.com also uses a similar license. This is from their terms of use: 

Quote

6.3

 

You are not permitted to:

 

(a)

 

sell or distribute any Content (modified or not) by themselves or in a texture pack, material, shader, scale modelling papers (pre-printed or digital), scrapbooking pack;

While textures.com explicitly allow using their textures in games in general, this section seems to be incompatible with CC-BY-NC-SA as CC in my understanding explicitly allows redistribution. But while trying to research this licensing issue I noticed that textures.com was used for assets at least in the past.

So how does this work? Are there exceptions where TDM assets can use a different license? Or is textures.com a relic of the past from when licensing wasn't enforced? 

If it's the first case, there would be other potential sources of assets that could be adapted for TDM. For example I'm a paid subscriber to blenderkit.com, a library with almost 2000 models which can be used under a license that doesn't allow redistribution and some of them probably could be used in TDM's setting. They're mostly hipoly with PBR materials, so retopology and some material work would have to be done, but that's less work than creating models from scratch and I want to try to do it anyway.

So what's the situation?

For those who are sensitive to motions like that as well, I've attached a patch.

I hope it or a variation of it will be accepted into the main The Dark Mod codebase, since it'll help those players.

The new cvar "pm_mantle_tilt_mod" is a value between 0.0 and 1.0. The default is 1.0, which means no change from the traditional tilt view angle. A value of 0.0 is no tilt. I personally prefer 0.1, which tilts the view only 10% of the traditional amount.

idCVar cv_pm_mantle_tilt_mod(
  "pm_mantle_tilt_mod",
  "1.0",
  CVAR_GAME | CVAR_FLOAT | CVAR_ARCHIVE,
  "The multiplier applied to the tilt view angle during a mantle",
  0.0f, 1.0f
);

(Patch based on r10036.)

Edit

The finalized version present in TDM 2.11.

"pm_mantle_roll_mod"  default:"1.0"                                                              
The multiplier applied to the roll view angle during a mantle

mantle-tilt-mod-v1-r10036.diff

I'm Herwig (30) from Germany. I'm a songwriter and Music producer who offers the Mod Team to write Ambient Music or Original Soundtrack for the Game or for some Missions.

If this help is needed please feel free to contact me here!

scriptEvent entity getAttachment(string attName);

Get the attached entity with the given attachment name Will be NULL if the name is invalid or if the entity no longer exists

Spawnclasses responding to this event: idActor

scriptEvent entity getAttachmentInd(float index);

Get the attached entity at the given index. Will be NULL if the index is invalid or the entity no longer exists

index: starts at 0

Spawnclasses responding to this event: idActor

scriptEvent float getNumAttachments();

Return the number of attachments on an AI. Used to iterate through the attachments if desired.

Spawnclasses responding to this event: idActor

But as you can see, they only apply to idActors.  I dug into the source code and found that the code that handles all 3 of these functions just calls into the idEntity base class to retrieve the attachment, and there does not appear to be anything specific to idActor with those.  As such, I'm fairly certain that these script events could be moved to idEntity without breaking any existing usages (assuming script event inheritance works as expected).  Are there any downsides to doing this? 

I can create a bugtracker entry and/or provide a patch if there are no objections.

This is a manifest to install the installer as a flatpak. It contains the following:

• com.thedarkmod.TheDarkMod.yml - Flatpak manifest, easily readable.
• com.thedarkmod.TheDarkMod.desktop - Desktop file. You can right click on it in an app launcher and choose "Installer" to open the installer program.
• com.thedarkmod.TheDarkMod.png - Icon for .desktop file, can't use .ico for this.
• launcher.sh - Simple script to initialize installation directory and choose between installer or game.

On first run the script will copy the installer to a sandboxed location under your user home (~/.var/app/com.thedarkmod.TheDarkMod/data/TheDarkMod), and then start it. The game will then be installed to this location. Afterwards, the script will check for the game launcher instead and run that if possible, otherwise it'll try to run the installer again. You can also open the installer by right-clicking on the icon and choosing "Installer", or manually by running:

flatpak run com.thedarkmod.TheDarkMod install

For the people using prime/optimus systems it should also default to running on the dedicated GPU rather than the internal when using the desktop-file.

To test this app locally you need to install flatpak-builder and the org.freedesktop.Sdk runtime in flatpak. You can then build and install the package by extracting the zip, entering the tdm directory in a terminal and running:

flatpak-builder --force-clean --user --install build com.thedarkmod.TheDarkMod.yml

To get this app published on flathub you can submit it via the instructions at https://github.com/flathub/flathub/wiki/App-Submission. I've never submitted an app myself so I don't know exactly how it's done, but AFAICT, it should just be putting these 4 files in a git repo and send a pull request to the flathub repo on github.

Feel free to edit any files as you see fit. The app name was based on the webpage URL, but maybe you want to change that or edit the capitalization or something like that.

thedarkmod-221003.zip

I've added "toggle crouch", which works like Thief, to my build. That is "toggle crouch" on key press. I've been playing that way since the end of March.

It's not as simple as changing it to "toggle crouch" on key press, because ladders and ropes need to be carefully considered. I'd imagine there may be differing opinions about how to handle those scenarios. I made it work as similar to New Dark as I could.

I've changed the code in my build to always work that way. In the Tech Support post, there was a request for an option to choose between new and old "toggle crouch" behavior, so I'll have to add that functionality to my patch. How shall I prepare my patch?

Some things that I found that I needed to consider:

• When on a ladder/rope, what happens when "toggle crouch" is pressed?
• When detaching from a ladder/rope, what happens when "toggle crouch" is held down after a slide?
• When attaching to a ladder/rope, what happens to the crouch/stand position?
• When on a ladder/rope, does "toggle crouch" crouch/stand or slide?

Here are the decisions that I made, which I hope is a good starting point for us to discuss what we all may want.

• Do NOT use crouch speed while climbing on a rope/ladder. (This is actually a separate issue / bug that I fixed.)
• While on a ladder/rope, "toggle crouch" behaves like regular crouch.
• While sliding, stay crouched until "toggle crouch" key is released. The player will stay crouched once detached from a ladder/rope and will stand (crouch off) on "toggle crouch" key release.
• When the player is crouched ("toggle crouch"), looking up, and moving up on a rope/ladder, change to standing position (crouch off). This makes it more likely for a player to be in a standing position after mantling from a rope/ladder. Note, on "looking up" is important, otherwise there would be a jarring, unexpected upwards movement on ladder/rope attachment when the player intends to go downwards.
• When the player is crouched ("toggle crouch"), attaches to a ladder, and is looking down, stay crouched.

It may be difficult to imagine how this works. Feeling it while playing makes a big difference.

Summary: I've been playing with my custom build since the end of March. Feels great. Feels like Thief/New Dark. I look forward to hearing your thoughts. How shall I prepare my patch?

Cheers

- CMake: Added check of compiler for support C++14 standard

CXX_STANDARD supported values are 98, 11, 14, 17 and 20.

Ref: https://cmake.org/cmake/help/latest/prop_tgt/CXX_STANDARD.html

- ThirdParty/devil: fix for MCST lcc compiler

LCC-Win32 and MCST lcc compilers define same identifier (LCC).
Have been added check of MCST Elbrus 2000 (e2k) architecture.

Ref: https://en.wikipedia.org/wiki/Elbrus_2000

Attached the patch with my changes.

C++14_check_for_CMake_+_fix_for_ThirdParty-devil.patch

• run TDM and TDM-updater.
• run TDM and TDM-updater with commandline arguments.

Normal version - More advanced version

• run TDM and TDM-updater.
• run TDM and TDM-updater with commandline arguments.
• Ideas: screen resolution tester, application logger, Mission remover (outside TDM), etc.

Download the work-in-progress-version of the lite or standard version from my google drive and extract the executable in your tdm-folder.
Sourcecode of these versions will be publisched on my github repo and google drive)

This launcher is inspired by darkloader, thief 4 setup and tdminstaller-win.
Are other things you want to add to this launcher?

Upgraded to 2.09, then messed with shaders some more. Mainly the glprogs/stages/interaction/ ... common & ambient, b/c they control bulk of lighting.

• broke out diffuse & specular lighting from each other again.. and figured out why specular was getting cut-off seams last time. It's because I didn't mul specular by light source, which takes attenuation into account. Did that this time.
• ambient.fs specular is generated from world light shining straight down, so was creating a spotlight under player on shiney floors (eg: gamma room floor on Training Mission). So, applied a 1-NdotL to have it use the light dot and not apply to surfaces perpendicular to the world light source (IE: ground).
• common & ambient .vs files didn't look like they were normalizing T,B,N before loading them into TBN matrix, so normalized them. (I think one was using a clamp, but switched it to normalize). Not sure it makes a difference or not.
• common.fs generated the normalized & TBN'ed "N" variable, but then used RawN in the NdotX calculations. Switched it over to using N instead. Not sure it makes a difference or not.
• ambient.fs was using AmbientRimColor to apply a fresnel highlight to final color, but after testing, didn't seem AmbientRimColor.a was ever > 0, so no fresnel effect was occuring. So, added an "else" branch to fallback on a generic fresnel highlight. (The velour-style fresnel I used in previous shader). Tweaked it to try not to look gaudy.
• Got rid of most of the fresnel effects I slapped in, or shifted them around a bit. The "eye adapt" was just half-arsed specular effect, so ditched it. The backlighting I just use on ambient.fs, but gets washed out a bit in darker levels and because specular is negated on floors in ambient.fs now. Still, it blends with fresnel to help out some. The velour fresnel was moved as alt fresnel effect in ambient.fs when AmbientRimColor not set or pitch black.
• switched ambient.fs to use halfdir Blinn-Phong specular like common.fs does.
• instead of just using generic default specular amount for things w/o specular texture maps, decided to BS specular maps for them by using the diffuse texture map.. did a dot(diffuse.rgb,0.33) to get diffuse lum and create a faux specular grayscale map which I then mul the default specular amount by to tone it.
• Also did that to create faux AO maps... initially used diffuse to do all AO maps, but noticed pillars in Training Mission looked awkward, b/c the AO was following the dark seams in the marble instead of the carved lines. So, if anything has a real specular texture map, used that for the AO map, and if it didn't, used the diffuse map as both faux specular and faux AO map. The faux AO blocks the ambient shadow light, but a final AOcolor is added to keep AO areas from being pitch black. This helps darken areas down some. Might be tacky in some places.

I've attached a .zip that contains interaction.ambient.vs, interaction.ambient.fs, interaction.common.vs & interaction.common.fs ... main files I farted around with.

Here's some screen shots.. noticed some issues...not sure if my messing around caused them or what...

ambient world light seems to always come straight down regardless of level designers placement of moon in sky.

This puts the ambient specular always underfoot like a spotlight. Without some kind of occlusion factor for it to know when player is indoors or not, dark dungeon floors get lit up with ambient world light. That's why I used a 1-NdotL to kill specular on floors. The ambient world light is still nice to make speculars shine on walls and objects. Just had to kill it on floors, like the one here in gamma room on Training mission. (This was done before I clamped specular color down to be a dull shine).

Some windows have a weird rectangle shape that doesn't blend well with the scene. It's like something that isn't controlled by ambient or common is not taking light attenuation into account. I experimented to see what ambient and common control, and there's some window textures that would stick out, which meant there's another shader being used to draw some textures. Couldn't seem to find it. I didn't spend much time digging.. but just looking in teh stages shaders, I didn't see anything else that could be doing that. The manylights shader doesn't seem to be used (?), so skipped messing with it. Not sure what's causing this.

Seeing how Faux AO map will impact things. It uses either specular map (if object has specular texture) or diffuse map otherwise. Basically just getting the dot luminance of either to create a single float value gray-scale to mul light dot by to add more darkness to things. A final AO color is added in to keep things from being pitch black.

Certain diffuse textures are missing .. something.. or processing is casuing this chunky issue. I noticed it intiially on some tile floors and windows on some levels.. in the distance the textures seemed to have a weird "acne" on them. But, when closing in, it would disappear. I thought it was specular. But, in testing fake AO map using diffuse, I noticed this. I have no clue what's causing it. A lot of these things blend into shadows, so don't notice them.. only reeally notice them in light (which is why I noticed them on windows and such). This one stuck out, b/c it's using the diffuse map as a fake AO grayscale map. Not sure what's causing it.

Fallback fresnel used when AmbientRimColor is 0 or not set. Tried to keep it subtle. It's more noticeable when moving around. Have to keep it subtle, b/c too much will make things in light appear ashy, and things in shadows look glowy. Too little, and it's as if it's not even there. Has to be just a little to help add depth to shadows. I figure enhancements are "death of a thousand" cuts... lots of little enhancements add up instead of one big honking one looking gaudy.

Fake AO. Most things don't have specular texture maps, so using diffuse map as fake AO means darker things darken up more.. hence the wall and grass are darker. Stone path is darkened in shadows more, too, though. With fresnel highlights and such going, the darkened areas still have depth.

This is a WIP. Just farting around, and trying to apply stuff I learned when dinking around with other game shaders, and experimenting to see what happens when I try different things.

glprogs.states.interaction.011.zip

"manual" install the darkmod.

(like unpack the updater in the right folder, run the updater, set the right permissions, create shortcuts, etc)

For these users I scripted some native graphical installers:

• Darkmod installer (windows), created with NSIS
• Darkmod installer (mac) (automaticly multi-language), created with iceberg
• Darkmod installer (crossplatform), created with izpack (darkmod forum topic)
• Darkmod software package - deb (linux debian/ubuntu/etc), created with dpkg-deb.
• Darkmod software package - rpm (redhat/centos/opensuse/etc), created with rpm-build.

I posted the also the sources, so the TDM-team and others can implement this in their installation procedure and improve the code.

I don’t have the time to improve/update/compile the installers for every released version myself.

So i’m hoping that someone in the TDM team (or someone else) would like to do that.

I would to hear what your thinking about it.

Status/bugs:

• Darkmod installer windows is ready to go.

  (successfully tested on windows 7/8, and runs with admin-rights).

• Darkmod installer mac, missing stuff like shortcuts, tdm binary not included) and the updater will not automaticly run after installation.
• Darkmod installer crossplatform missing stuff like shortcuts, tdm binary not included) and the updater will not automaticly run after installation.
• Darkmod software package deb, running with the package manager displays some warning messages. but it can be ignored.

Is there a need for another mirror to host missions and game files at this time?

It's just a rough sketch, but hopefully it portrays the idea I have in mind! More data and filters could be added, but the idea in any case is that players (especially new ones) have an easier time finding missions suited to their taste.

The "mission finder" would be a web application, which could be placed either on a new domain or a subdomain of thedarkmod.com.

I have plenty of experience working as a data systems developer, so I could fully take care of this.

What do you think? Could this be useful? Maybe not as useful as I think and would be a waste of time instead?

Interested in hearing other people's opinions on the subject.

Edit: Maybe this should have been posted in "I want to help"? Please feel free to move this thread there