One Minute Game Review by The Happy Friar:
Also on Rumble:

idTech 4 (aka Doom 3 tech) Discord Server!

Main Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - bkt

I'd guess they took their New Order code base and updated it.

Likely the same for Evil Within 2?
Just going to leave a more reliable link to the story:

id Tech 6 General Talk / Re: Update 5
December 14, 2016, 11:56:09 AM
Quote from: VGames on December 13, 2016, 10:02:59 AM
What's to laugh at. Raising awareness would be the first step.

Oh wait a minute. Are you an anti-Trumper?
Are you a pro-Trumpeter?
Quote from: bitterman on September 29, 2016, 10:33:22 PMSo if I needed to activate trigger with brush/model (any "non-living" entity) then I shoud bind func_activator to these entities.



I used the func_activator for the following sequence (the video was recorded well over a year before release, so please ignore the placeholder voice work):

As the player could move around on the platform I used an activator to open the doors based on the platforms position.  This meant that if the player was standing at the very back of the platform, the door would still open at the correct time without any chance of the platform clipping into the door.
Quote from: motorsep on September 13, 2016, 08:25:30 AM
Folks, do yiu realize you don't have to follow terrain closely?!

And by that I mean you can use regular brushes (bottom is caulk and the rest AASsomething) and crudely block out terrain.

The AI only sees the AAS sectors and will blindly collide with the terrain.  It's not a matter of 'Doom 3 doesn't support terrain', it does.  It's just a matter of working within the boundaries of its implementation. 

There's no gray area for whether or not an AI will be able to navigate terrain.  Consider that if the AI only see's the valid AAS areas (generated from brushwork), then you simply need to make sure that the terrain represents this valid space.  There will be a certain angle at which an AI/Player can't walk up and if any tri's in the mesh exceed said angle, the AI won't be able to navigate correctly.  So, based on that simple logic, if you create geometry that matches the logic in terms of playable space, you'll have completely valid terrain.

Seriously though, enough with the conjecture.  If the examples given by those of us who have worked with aspect of the engine isn't enough, then go and dissect an example from Doom 3.  Open maps/game/ and work it out.
Quote from: bitterman on September 10, 2016, 10:49:57 PM
Guys, your videos are good, but this is more advertisement than a tutorial on creating terrain. I had a thousand questions, how it is implemented, but who is going to answer them?  ;)
You're not getting a tutorial from me.  Why not jump into the editor, find an example from the Doom3 campaign that works as intended then reverse engineer it, instead of waiting for step by step instructions from those who've done that very step.

Everything you need has been explained here, simply connect the dots.
Here is an example area from False Dawn. The entire section of the map up to the elevator is comprised of a static mesh floor.  Caulk brushes are underneath it and generate the AAS allowing the AI to completely navigate around the hallways as if walking on flat brushes.

(The russian stream playthrough is the best quality commentary-free playthrough I've found).
Quote from: motorsep on September 08, 2016, 12:26:48 AM
Quote from: bitterman on September 08, 2016, 12:01:54 AM
More info:

Here is proof it just works (and that map didn't have AAS compiled at all):
Showing a video that has no AAS compiled is useless as an example.  Without AAS the AI can't see anything in the environment and won't be able to navigate any obstacles, meaning the AI's simply honing in on the players position blindly.

For bitterman:  caulk underneath the static mesh, the AI will see that and navigate over it using the AAS areas generated.  So long as the mesh doesn't exceed an angle the AI can't navigate, then they will be able to navigate effectively.
Quote from: motorsep on September 07, 2016, 04:24:44 PM
Quote from: BielBdeLuna on September 07, 2016, 02:55:43 PM
it might not be as trivial or easy as this, specially if the terrain isn't flat which should be the point of making any terrain with a mesh.

It's is as trivial as he described and works out of the box.


Just put caulk/clip under the terrain and it'll generate AAS that the AI can use to navigate over the mesh.  The AI will see the AAS zones generated by the flat brushes while they will collide with the mesh.
I'm trying to remember what scenarios I used it in other than for model collision... because there was a collision material for models too...

I don't have the False Dawn source files handy to check against.  Though caulk is still the most valuable of the common materials.  The majority of that project was bright pink in the editor.
I can't remember exactly what the actual differences were between nodraw and caulk.  However caulk is generally used as a non-rendered seal for the map.  eg. place it behind patches/meshes to create a sealed room, but the engine won't process any rendering for the caulk surfaces. 

As for nodraw, all I remember is that I used it a lot for any invisible collision on meshes.  If memory serves, it doesn't generate AAS.

So, how about you put together a testmap to work it out if you're not sure.  See what their collision/draw/culling values are like.

Strutt your stuff! / Re: fhDOOM (modernized idTech4)
August 30, 2016, 11:32:10 AM
To be honest, I'll take my engine mods simply as engine mods.  Bundling gameplay mods in just muddies the waters, especially if there aren't easy toggles to revert the game back to its original gameplay state.

One of my biggest issues with Sikkmod for example was the way that it modified so many small aspects of the game loop.  If I wanted to simply replay the game, or try a custom map, it was a pain to relive that same experience with added visual flare.

In saying that, I'm aware that I completely reworked a lot of Quake 4's features in my last Sikkmod powered release, but that was kind of the idea.

fhDoom sounds like it would be a great starting place for someone to either build a map, mod or just play through existing content with a nice coat of paint.
Strutt your stuff! / Re: Perfected Doom 3
July 07, 2016, 05:23:31 PM
I'd echo what EoceneMiacid said.  I couldn't really tell what the mod is showcasing based on the V7 trailer.  Sure, there was action and some weapon mods, but it was mostly explosions and gunfire in familiar scenes with slightly different looking enemies?  It also seemed odd to me to spend so much time on cutscenes from the original game.  As a potential customer, I'd much rather see what makes your mod stand out featured front and center.  Sure there are features listed on the ModDB page, but the trailer left me scratching my head and not too enthused to follow through to find out more.

You definitely put some time into making a fun trailer that syncs up with your chosen track, but as I mentioned above, it'd be good to see a more focused effort to get my hyped next time.  Just a bit of honest feedback for you.
One thing worth noting about Sikkmod is that -- outside of the gameplay modifications -- is essentially a collection of post process effects.  If someone was to integrate the same effects into any branch at the engine level, it would be essentially the same. 

Something very useful that Sikk did was making sure that all of the variables for each effect were visible as cvar's.  He also implement multiple solutions for various effects (eg. Crysis SSAO vs HBAO or Tone mapping vs UE3/Cryengine 2 style LUT colorgrading).  In the case of working with him and a custom build of Sikkmod on my last mod, I was able to customize the visual style of the mod while balancing the quality of various effects for performance (eg. Bilinear DoF or Bokeh DoF).

The majority of the implementation of the various effects were simply standards available online that he'd managed to integrate in an intelligent (and versatile) way.

The custom version he developed with me for False Dawn had a number of very cool additions not available in Sikkmod V1.2 for D3.  Sadly he went MIA right before I completed the project and got the final source off him :(