Search the Community

https://www.gamesradar.com/witcher-3-mod-uses-ai-to-create-new-voice-lines-without-geralts-original-voice-actor/ https://www.ibtimes.com/witcher-3-story-mod-stirs-controversy-over-ai-generated-voice-acting-3237250 https://www.inputmag.com/gaming/video-game-voice-ai-human-actors-witcher-3-mod-controversy My position: get over it, voice actors. The writing has been on the wall since Vocaloid was released. I think we've seen games shipping with 10 to 30 gigabytes of voice data at this point. You could imagine some game using procedurally generated text (like GPT-3 used to fuel NPC chatbots), team written scripts, or crowdsourced scripts to get to the equivalent of 1 terabyte or more of lines. Powerful 8-core CPUs are becoming the minimum standard for gaming, and between the CPU and GPU there will be more than enough computational resources available to synthesize voice lines and other sounds, like object collisions, in real time. Assuming 500,000 pages per gigabyte, 1.5 minutes to voice a page, you can get 1.4 years of text-to-speech from 1 gigabyte of scripts. The legal issues are legitimate. I have no doubt that a court would side with voice actors who are having their voices "stolen", citing personality rights. At the same time, companies could figure out how to mix the voice samples of hundreds of real people together in training data, and adjust parameters to come up with an infinite number of indistinct voices that can be used without paying anyone. Meanwhile, fan efforts can rip off real voices from Hollywood and voice actors, or specific voice performances like Stephen Russell as Garrett. It's not worth it to sue them, and they can come together pseudo-anonymously and distribute code using torrent sites if needed. In some cases, an amateur will do the voice acting and then a different voice style will be pasted over the original recording. We could also see pure text-to-speech with a markup language to add emphasis, vocal cadence, etc. In either case, an algorithm can definitely transfer or fake the "breathing" and pauses. It's also likely to be considered art. Ignoring the fact that an "invisible sculpture" can be considered art, there will likely be a lot of creativity or at least fine tuning when writing a script, working with an AI that generates scripts, and perfecting the voices.

Since upgrading my gpu to an nVidia RTX 3060, I can only set v-sync via driver setting Sync to Vblank. At least I thought that's what that setting does. Enabling V-sync in TDM does nothing. I use Manjaro Linux xfce Both TDM 2.11 and 2.12

I am curious as to which distributions and tool version are known to work when compiling either the 2.10 source archive or svn trunk. Cmake errors out with both Archlinux and Mint 21 for me. For good measure I will include some information I gathered about these errors, but I just want to be able to successfully build tdm on something. (My end goal is to compile in either pulseaudio or pipewire support for openal if that's possible; even though this thread got my audio working (arch audio subsystem is pipewire{,-alsa,-audio,-pulse} & jack2 ) ) I've also tried compiling against the latest svn trunk, that fails too, although the point in which it fails is slightly different. What do the developers use to build the linux release? System information for the failed builds against the 2.10 archive: System #1: cmake 3.22.1 | gcc 11.3.0 | kernel 5.15.0-56-generic | distro Linux Mint 21 System #2: cmake 3.25.1 | gcc 12.2.0 | kernel 5.15.83-1-lts | distro Arch Linux src archive used: sha256sum thedarkmod.2.10.src.7z 73aa974635293e6ca07396be19901355f8224637bdf3ce73404b8eef74148a1c thedarkmod.2.10.src.7z Build command from root of extracted src: [ -d build ] && rm -rf build mkdir build && cd build && cmake --debug-output --loglevel=DEBUG -DCMAKE_BUILD_TYPE="Release" .. &> cmake.$distro.log && \ make -d --debug=a -j &> make.$distro.log ; cd .. make.mint.log.gz make.arch.log.gz cmake.mint.log.gz cmake.arch.log.gz

Hello, I have some roles to distribute for: - builders - mages - pagans - inventors - regular guys (you know which type of guys, the bad watchmen guys) Any new voice talents around?

Voice acting?

After all the research I've done so far and comparisons, including reading the history of the development of TDM. It seems TDM would be the best engine possible for DOOM 3. Any thoughts, or avenues to pursue on this from anyone constructive would be much appreciated.

There's a bit to unpack here, but I'll ask that you please read this post and my motivation before giving a verdict. At the moment I'm just opening this thread for debate: If the idea isn't shot down as absurd and we agree it's possible to consider, I may proceed with a bug tracker entry otherwise forget I said anything. Also this post does NOT imply we have developers willing to work on it, if no one in the team considers it an effort worth investing time in I'm not going to beg. It would be a major change that takes time to test, so even if it does happen it would likely be next year at earliest (TDM 2.11 or later). Lastly this is intended solely as an OPTION... as the person suggesting it I don't believe it should ever be enabled by default due to how it would alter the gameplay, only offered as an extra component for players who want the mechanic. That being said, the idea occurred to me yesterday and I've been excited thinking about how it could work, it actually kept me up last night and I'm getting this off my chest to hear some feedback so I can fall asleep more easily tonight Let's get into it. The idea is partly inspired from some horror games I watched Youtubers play a while back. They had a mechanic where the engine accessed the player's microphone to read the volume: If you made too much noise, monsters would hear you and come after you. The player had to stay silent else they'd be in trouble. Last night it clicked with me how perfectly this mechanic would fit into TDM with our AI and gameplay. Imagine if any noise you made IRL risked alerting nearby guards... whereas oppositely you could use your voice to distract them to your location and move past. I'll discuss both uses below, first let's go into some technical aspects. This would obviously require engine changes: The engine needs to learn how to access the default microphone device from the sound system, enough translate the volume into a to 0 - 1 float. The game reads this volume at an interval balanced between accuracy and performance (eg: 0.5 seconds). Based on the loudness it captures that call, the same signal emitted by sound sources to alert AI is propagated from the player's view origin. The effect tapers off with distance in a fashion similar to s_mindistance (full effect below this radius) and s_maxdistance (fully ignored beyond that radius), reasonable defaults might be min 4 / max 16 measuring with the radius of a speaker in DarkRadiant. Multiplied with that distance and the AI's acuity factor, the mic volume determines the alert level being generated. The ranges I'm suggesting for this would be: 0% to 20%: Ignored, doesn't generate any alert. Ambient noise and insignificant sounds are usually in this range for all microphones. 20% to 40%: Can go up to the 1st alert level, makes the AI mumble and say "did I hear something" but move on and ignore you. 40% to 60%: Can produce 2nd alert level in guards, making them stop in place to look around. 60% to 80%: 3rd alert level allowed, guards realize something is up and start searching the area. 80% to 100%: Loud enough to make your presence in the area known, 4th alert level is possible making guards draw their weapon and dare you to come out. 100%: So loud that your exact location is clear to the guard as if you bumped into them, 5th alert level activates and the AI proceeds to attack you. So what is the point of this? Some would see it as an annoyance and the player sabotaging themselves, though just like selecting a higher difficulty level it's a controlled form of making your run more interesting. I for one am in it for the immersion and ingenuity: It would offer a new way to integrate the player into the world, using hardware almost every player has: A headset with a microphone, a mic on their desk, even one in their webcam. This has been done in small games but never attempted in a full stealth game for my knowledge, TDM could be among the first project of this genre to use the idea which would be an extra selling point to make it more interesting and appealing. The first aspect is fear of being heard. Imagine you really are the player in a typical FM, hiding in the shadows as guards are walking past or running around looking for you. The last thing you want to do in such a circumstance is make any noise: You'd be careful about your breathing, and god forbid you need to cough or sneeze. This mechanic would emulate just that: Players have to be aware of their breath as an enemy walks past them... if for any reason you feel like sneezing or coughing, you better hold off or the guards will become alert in-game. To me this sounds super neat... I can already imagine playing TDM with the added fear and adrenaline The second part is how this can be used to attract guards in a controlled way. Jumping in place works but it's hard to make guards come and investigate, with this you'd do it using your voice. The fun here is learning how loud to speak to get the desired alert level: If you whisper to the guards they won't hear you and just walk past, but if you scream too loudly they'll know where you are and attack you on sight! You'll want to say something like "hey" at just the right volume to make them investigate. As a bonus to end the idea, why not introduce this for friendly AI too as a fun detail? You know how some FM's have pubs the player can enter where the maid and party-goers won't attack you; Whenever the player says something in their mic, friendlies and neutrals look at you and use the generic response bark! Imagine how much fun we'll have taking breaks from the objective to say all kinds of stupid things in the mic and each time someone responds with a "hello there" or "you said it" or "yeah yeah"

I'm not sure if this should go here or in the Tech Support section, so If this is the wrong place, please let me know. As a hobbyist game developer, I'm curious about how the Light Awareness System works in The Dark Mod. I've been poking through the source code, but haven't found any definitive math explaining it, or code comments mentioning on how it's calculated. If anyone could point me to exactly where to look, or could explain what steps and/or math it's using to calculate how visible the player is, I'd greatly appreciate the insight. For reference, I'm trying to implement a similar system in my own stealth game I'm working on in Unreal Engine 4.

I still need a second NPC voice actor for my upcoming mission. It's about a person who is stuck in the level and will have all sorts of things to report over a loudspeaker system and who is between scared, desperate and hysterical but all with a funny undertone. So things like: "How could that have happened to me? Now, of all times, when I only have to fulfill another 35 months of my three-year contract! Any new or old volunteers here? Andros (Jonus Thorunsson, an Inventor) and Malasdair (The Thief) are already on board.

I wouldn't normally do it, but I put a member on my ignore list because the posts were upsetting. Very upsetting. I enjoy TDM and DR and enjoy the forum, except for the one person. Usually I check the forum out as 'guest' but will auto sign-in so there's no trace of the ignored person's post, mentions, and all of it. But it doesn't work. And I'm outa here if I have to suffer any more of that.

It doesn't work at all, v2.09a and the latest dev build. If I input eg. "r_showShadows 2" nothing happens, then when I input "r_showShadows" it gives the info that the value is currently set at "2", that default is "0", along with info on how it should work.

I'm in the process of creating WS6 (Baleford Museum) and WS7 (The Builders' Forge). I hope to release them simultaneously in the next 6 months, health permitting. I'm at the point where it's time to install the voices for a few conversations and monologues. If you're interested in providing some voice acting, please let me know. Experience isn't necessary (new voices are ALWAYS welcome), and I'll walk the inexperienced through the process. Just make sure you've got a good setup (so we don't need to manually filter out background noises, for example). I won't be doing any post-processing on your voices; what you say is what we'll hear. I'll be managing the ogg creation and sound shaders and folder management, so all you need to worry about is your performance. I have some of the scripts finished, so we can dig right in as soon as you're available. Thanks!

Hello dear developers! Why do many variables not work in your darkmod.cfg? I changed the following variables and set the value to " 0" seta pm_bobroll "0" seta pm_bobpitch "0" seta pm_bobup "0" seta pm_runroll "0" seta pm_runpitch "0" seta pm_runbob "0" seta pm_walkbob "0" seta pm_crouchbob "0" I launch the mod. . I run training mission. And all settings are reset!! The game also does not read variables from autoexec. cfg.

It’s my first FM I’ve ever created for any game. I’ve essentially finished the map but I’d like to have some feedback. I’m still open to changing things depending on feedback, so it hasn’t officially hit the beta stage yet. The mission is quite different than the standard thief/dark mod mission (more story driven and atmospheric) so I’ve got no idea how well it will be received. Due to the nature of the mission it will have lower replayability so keep that in mind if you wanted to test it before its official release. When there’s a few people interested I will post it in the beta testing forum. Thanks. Type of mission: Mansion/Estate/Horror Some screenshots attached.

I get the feeling antialiasing does not wish to work for me, even when I use the 4x or 16x option in the main menu. I remember getting the impression that it worked sometimes in the past, but yesterday I looked at the edges of some objects and could clearly see the jagged edges of the geometry. I wonder if this is specific to the Linux version: I have an AMD card and use the free amdgpu driver... there was talk about Mesa having issues handling AA properly depending on how the engine calls it. What do you suggest I check?