faatal

They should be able to just like zombies can fight animals. I don't know. Still a thing. We have a programmer working on it full time.

Sure, I've looked at it, but it works about as well as Vulkan in Unity. Unity is supposedly making a push to improve DX12 support. I saw it on their roadmap. Maybe next year.

I'm not against decision trees for any reason other than I don't need them for what I want to do. What we have now could be called decision lists, which are lists of tasks that get checked in order. Bandits will probably have more task types added to support new behaviors. The key thing I'm doing now is a full cleanup of the zombie controller parameters, states and transitions and their avatar controller code, to get ready to be copied for the bandits, as I don't want the bandits starting from some legacy mess.

The number of bandits that are spawned at any one time. Say a bandit takes x2 the performance of a zombie, then I would not want more that half of them in the world compared to the number of zombies a computer can manage and still have a reasonable FPS. Now that does not mean the world will have bandits everywhere. That depends on game play goals. I would like bandits to be rare, because this is a zombie game and 99% of the population is dead or a zombie.

EAI has been optimized over the years I've been working on it and UAI being dead, I have not touched it. I do plan to remove it at some point, but hopefully the bandit framework will be adequate to extend from for modders. Good bandit performance and numbers is a goal, so they will run adequately on all systems. It is still in the works. There were some issues with platform changes and then there was a bad merge of some of my changes that they backed out and may end up having to hand merge.

Optimizations are not really in progress. They just happen occasionally as I or someone else sees something to optimize. It happens throughout the development process. Min specs should not be changing from what they are now, but we just recently increased the CPU core/frequency specs. Min spec means you can run the game, but not necessarily every feature or mod well or at all. Dynamic mesh is a good example. I would not be using it at min spec, but it is not something we enforce. Nav is 6k. Thanks for playing. Not decided. Will depend on how it impacts game play.

Bandits kill squads could be as simple as a blood moon, but instead with a set group of bandits at whatever time it decided to spawn. We will have to wait and see. Probably. Should be higher as bandits will be more rare.

Bandits of the same faction will have some type of coordination.

There should be melee and ranged bandits. Some may do both. Melee are the first ones I'm working on. They will be hostile. Possibly later there could be neutral or friendly ones belonging to a faction. They will not loot stuff or destroy things unless they are in their way just like the AI currently does. They are zombies with weapons, but with different behavior. Like they will hide in cover and wait to ambush you.

It is a model an artist is working on, so finalish.

Not likely. That would probably have to go through Unity and I have not seen Unity mention it. More time seems to be spent processing data than loading it. I was just saying today, we need more improvements on processing data, because if we had infinite data read speeds, load times would probably still not be much faster.

I don't know what the average POI designer typically does for rooms, but they should already be able to include the wall inside the volume. Now volumes are not meant to include the full 1m wall thickness plus all of where you can stand beyond them (just a little) as we don't usually want you beating on walls and waking up the volume.

1 Depends on performance. Currently water is not infinite, so it would drain the higher portion. 2 I'd would guess they will add some as they improve POIs. 3 There is always a chance. 4 Not that I know 5 Should be wandering and inside POIs

Well, the only feature I can confirm about a21 right now is, we won't release it until it has some new features. I would like to have more vehicle choices. I saw a picture of a sweet new police car model today that made me want to drive it, but I'm currently not allotting time to it for a21. In the end it could happen if I find some time and the art is setup for it.

Water is making good progress, so there is a good chance it will be in A21.

Normal stealth checks would not be desirable in the attack volume case as the zeds are often around corners, enclosed or facing the wrong direction and the player may be making little noise which means default unskilled stealth would mostly work, yet the designers want the zeds to attack. The one time check is to allow stealth to be useful as a skill you can put points into and allowing for variation in the way you entered the volume compared to zed locations. The attack uses SetAttackTarget, which is a standard way of the tracking you and used in cases like when you hit them or they see you. I am reducing the trigger's call of SetAttackTarget from 60 to 20 seconds, which seems more reasonable for the attack to time out, since it refreshes anyway if they see you.

The person who made the video did not notice he is in broad daylight when he tested the attack volumes, so yes they saw him as his stealth was horrible. Not having shadows on also makes it hard to know when you are in sun or moon light! They do the normal stealth checks and are actually worse at detecting you since they can only hear you and not see you. A sleeper in the open could see and hear you.

There are two types of stealth checks that happen. 1 Continuous - A player stealth update calcs their light level, updates noises and sets themself as noise source on AI entities if they are the higher than other player's noise. AI checks players in their vision cone based on distance vs that player light and then target the player. The player noise on the AI, if above a threshold, will wake the AI. Noise will also make it investigate and do a vision check, which could cause it to see a closer player than the one that made the noise. 2 Triggering a sleeper volume - This happens once per AI based on stealth modified light level vs distance and only determines attack vs active. They will go active in any event and continuous stealth checks may hear or see you at any time after that.

I'm pretty sure there is a MP weather sync bug, but have yet to track it down.

The stealth check is in there. It is per zombie.

I spent many hour this weekend fixing some block entity culling issues. It should be greatly improved. Sleeper volumes are 3d, so they are already managed vertically.

We do as needed or if people choose to. There were 11 changes to SVN today by 6 different people and many are watching streamers, testing and making bug tickets. I do at normal times, but releasing is not a normal time.

Sure we do, if it is a major bug. It may not get released that same day, since it needs testing by others, but it would get worked on. Just like I saw performance issues today in downtown areas and debugged it and made some improvements. Needs more work, so I'll continue tomorrow, to hopefully go in a Sunday build to be ready for Monday.

Actually, lots of games take many years and tons of alphas/betas, you just don't get to see it, because they are not early access. I don't care how long a game took to make, as long as I got my money's worth. I buy plenty of normal games and have no clue how long they took to make. I buy some early access games and will sometimes play them right away if it looks far enough to be a good experience like I did with 7dtd years before I was an employee and I certainly got my money's worth. Other times I just wait for them to finish and then play to get the full experience like I did with Black Mesa. Now I could have just bought it when it was done, but it was a deal on sale, so I took the chance and just purchased it and then ended up waiting several years.

Experimental is basically a buggy version. Some features may be added before stable, but generally nothing major, since the goal of exp is to fix bugs and tweak before a stable release.